Proceedings of the Inception Workshop on the GISAIA Project 19 September 2013

Transcription

1 Inception Workshop on the GISAIA Project Sunbird Capital Hotel Lilongwe Malawi Guiding investments in Sustainable Agricultural Intensification In Africa (GISAIA) Proceedings October 2013

2 TABLE OF CONTENTS TABLE OF CONTENTS... i List of Figures... i ANNEX 1: Full Papers... ii ANNEX 2: Speeches... ii ANNEX 3: Worskhop Programme & List of participant... ii ACRONYMS AND ABBREVIATIONS... iii PREFACE... iv ACKNOWLEDGEMENTS...v 1.0 INTRODUCTION The Context Workshop participants WORKSHOP PROCEEDINGS Opening ceremony WORKSHOP OBJECTIVES Broad objective of the workshop Specific objectives of the workshop SUMMARIES OF PRESENTATIONS The role of FISP in the Malawi Economy: Any Prospects of Graduation? By Prof. E.Chirwa What are the farm-level impacts of Malawi s farm input subsidy program? A critical review by Dr. Rodney Lunduk, IIED, United Kingdom An overview of the Guiding Investing in sustainable Agricultural Intensification in Malawi by Jacob Ricker-Gilbert, Purdue University Role of Private Input Suppliers in Smallholder Agriculture: Insights from Malawi by Charles Jumbe, CARD Improving Maize Productivity through the Promotion of Sustainable and Profitable Use of Fertilizer: The Role of Complementary and Restorative Management Practices by Jonathan Kanyamuka, Bunda College, LUANAR Profitability of Fertilizer Use: Evidence from Malawi, Mr. Francis Darko, Purdue University Farm Household Efficiency and Farm Input Subsidy: Evidence from Malawi, by Mr. Francis Darko, Purdue University CONCLUSION AND FOLLOW UP ACTIONS Official Closing...7 List of Figures Figure 2: Understanding Household and Local Economy Impacts of Input Subsidies... 9 Figure 3: Budgeted and Actual Programme Costs Figure 4 : GDP Growth, Agricultural Growth, Poverty and Inflation, Figure 5 : Maize prices and Estimated Quantity Consumed Per Capita, Figure 6: Average Maize Prices, Tobacco Prices and Ganyu Wages Centre for Agric. Res. & Dev, Malawi i

3 Figure 7: Maize production and prices, Malawi 1991 to Figure 8: Factors affecting household access and supply of inorganic fertilizers by private input suppliers (Hypothesized) ANNEX 1: Full Papers Annex 1.1 Background Paper: The role of FISP in the Malawi Economy: Any Prospects of Graduation?... 8 Annex 1.2: BACKGROUND PAPER 2: WHAT ARE THE FARM-LEVEL IMPACTS OF MALAWI S FARM INPUT SUBSIDY PROGRAM? A CRITICAL REVIEW Annex 1.3: The Role of Private Input Suppliers in Smallholder Agriculture: Insights from Malawi Annex 1.4: Improving maize productivity through the promotion of sustainable and profitable use of fertilizer: the role of complementary, restorative and management practices Annex 1.5: Profitability of Fertilizer Use: Evidence from Malawi Annex 1.6: Farm Household Efficiency and Farm Input Subsidy: Evidence from Malawi ANNEX 2: Speeches Annex 2.1: Opening Speech by the Minister of Agriculture and Food Security Dr James T. Munthali Annex 2.2: Remarks by the representative of the Vice Chancellor Dr Kenneth Wiyo Annex 2. 3: Remarks by the Acting Director of Centre for Agriculture Research and Development (CARD) ANNEX 3: Workshop Programme & List of participant A Annex 3 1: Workshop Program Annex 3 2: List of Participants Centre for Agric. Res. & Dev, Malawi ii

4 ACRONYMS AND ABBREVIATIONS ASWAp : Agricultural Sector Wide Approach CA : Conservation Agriculture CARD : Centre for Agricultural Research and Development CNFA : Citizens Network for Foreign Affairs FISP : farm Input Subsidy Program GDP : Gross Domestic Product GISAIA : Guiding Investments in Sustainable Agricultural Intensification in Africa GoM : Government of Malawi GPS : Global Positioning System HIV : Human Immunodeficiency Virus ISFM : Integrated Soil Fertility Management Kg : Kilogram LUANAR : Lilongwe University of Agriculture and Natural Resources MAISP : Malawi Agricultural Input Subsidy Programme MEPD : Ministry of Economic Planning and Development MNL : Multinomial Logit model MoAFS : Ministry of Agriculture and Food Security MoAFS : Ministry of Agriculture and Food Security MSU : Michigan State University Mt : Metric tons MWK : Malawi Kwacha NASFAM : National Smallholder Farmers Association of Malawi NGOs : Non-Governmental Organizations NSO : National Statistical Office RSA : Republic of South Africa RUMARK : Rural Agricultural Marketing Program SACA : Smallholder Agricultural Credit Administration SFFRFM : Smallholder Farmers Fertilizer Revolving Fund of Malawi SPS : Starter Pack Scheme SSA : Sub Saharan Africa SSA : Sub-Saharan Africa TIP : Target Input Program US$ : United States Dollar Centre for Agric. Res. & Dev, Malawi iii

5 PREFACE The Farm Input Subsidy Program (FISP) was instituted in Malawi in 2005/6 growing season following a pervasive hunger crisis and chronic food shortages in 2001/2 and 2004/5 growing seasons. In 2004/5 growing season alone, about 4 million households were in need of food aid. Since then, government and development partners have been committing huge sums of income in making sure that citizens are food secured. Despite all the effort rendered by the government and all stakeholders in agriculture sector, reports shows that 60% of the smallholder farmers are net buyers and only 10% are net sellers. This has been attributed to low productivity of maize to most of the smallholder farmers. On the contrary, inter year prices have been persistently low and volatile making it difficult for potential surplus producers to invest in maize production, in a way affecting the performance of the private sector. However, it has been strongly argued that proper designing, implementation and management of the program ensures its effectiveness, efficiency and sustainability which also calls for medium term plan as opposed to ad hoc approaches which are not predictable With nine years of implementation in Malawi, FISP seems to receive different reaction from stakeholders as such Guiding Investment for Sustainable Agriculture Intensification in Africa) organised an inception workshop with an intention of soliciting views from stakeholders on how to make fertilizer subsidy more profitable and sustainable for smallholders in Malawi. The Lilongwe University of Agriculture and Natural Resources (LUANAR) through Centre for Agricultural Research and Development (CARD) would like to thank all those that took part in various ways in order to make this GASAIA workshop a success. In particular, CARD would like to thank the Ministry of Agriculture and Food Security for their tireless support and the Minister of Agriculture and Food Security, Dr James T. Munthali for the official opening of the workshop. CARD would also like to extend our gratitude to all institutional and individual stakeholders for supporting the Government in promotion of research on FISP that will assist in redesigning the programme to suit its intended goals. Targeting of beneficiaries which varies every year across space in Malawi, is vital in marrying the programs objectives and its corresponding output. We believe that allocating the limited resources (financial) to a more productive people will surely bring out obvious results making persistent hunger crisis and chronic food shortages history in our mother Malawi. CARD would like to sincerely thank Michigan State University and Purdue University for the collaboration hoping to continue carrying out research together. To all the presenters, we would like to extend our heartfelt gratitude for making the workshop colourful and all participants Thank you for the constructive comments and thoughts provoking questions we are sure that without you the workshop couldn t have achieved its objective. Centre for Agric. Res. & Dev, Malawi iv

6 ACKNOWLEDGEMENTS The Lilongwe University of Agriculture and Natural Resources (LUANAR) through the Centre for Agriculture Research and Development (CARD) at Bunda College would like to thank all participants at the inception workshop of the GISAIA project that will run from September 2013 to August In particular, CARD would like to acknowledge the technical support and collaboration received from the Ministries of Agriculture and Food Security and Economic Planning and Development who constitute that technical committee for the project. We would also like to acknowledge the resource persons who presented background papers to gain a better understanding of the design, implementation and impact of the Farm Inputs Subsidy Program. The papers presented at the inception workshop situated the research proposals presented by the prospective students to be supported by the project during the 2013/2014 and 2015/2016 academic years at Bunda College. The students are thanks for preparing and presenting the research concepts but most importantly, gratitude is due to all that commented on the proposals. The comments will help the students to conduct relevant research that will feed into the design of future subsidy programs. The Seminar would not have taken place without the generous financial support from the Bill and Melinda Gates Foundation in the United States of America. Furthermore, CARD would like to its collaborators for this project, Purdue University, Michigan State University for championing the preparation of the project that was submitted to the Bill Gates Foundation. As a new University, we are proud to be associated with this reputable research institutes and it is our hope that we will forge ahead in making serious commitments to collaborative research. Finally, the Organising Committee comprising Officials from CARD and the Ministries of Agriculture and Food Security and Economic Planning and Development led by Dr Charles Jumbe is being thanked for ably organising the workshop. May God bless you all for your individual and institutional contributions towards the success of the workshop. Alexander Kalimbira, PhD. Acting Director, Centre for Agriculture Research and Development Lilongwe University of Agriculture and Natural Resources Centre for Agric. Res. & Dev, Malawi v

7 1.0 INTRODUCTION 1.1 The Context The Farm Input Subsidy program has been implemented since 2005/6 growing seasons following two episodes i.e. the chronic food shortages and pervasive hunger crisis in 2001/2 and 2004/5 growing seasons respectively. These two episodes left 3.2 million and 4 million people in emergency of food aid. The government devised this programme as a silver blue solution to the food insecurity disease in the country following the argument that it is expensive to import than to produce for one self especially if the country experiences forex problem like Malawi. High population growth has led to small land holding sizes estimated at less than 1 hectare per household; hence increasing production by area expansion is not a viable option for Malawi at the mean time rather productivity is the only option. Smallholder farmers constitute a larger population of the total farming population for Malawi. They are characterised by small land holding size, low productivity, less managerial skills among others. The target of FISP is to ensure food self-sufficient both at the household and national levels. The food security status at a national level hinges on increasing productivity to meet consumption levels of the growing population estimated at 2.1 million metric tonnes per annum. Researchers have argued that low maize prices on its own it s a disincentive to the private sector participation adding that high maize prices could have forced maize deficit farmers to grow as much maize as they could on their piece of land even if they couldn t afford improved seeds and fertilizers. 1.2 Workshop participants The participants of the workshop were the key stakeholders of the Agriculture sector. These stakeholders includes government Ministries and Departments; Development Partners i.e. the World Bank, the private sector representatives, the academia, Research collaborators from Purdue University, members of press among others. The full list of participants is provided by Annex WORKSHOP PROCEEDINGS 2.1 Opening ceremony The Guest of Honour Minister of Agriculture and Food Security Dr James T. Munthali started by thanking the Donor community, private sector representatives, development partners, Research collaborators from Purdue University and Michigan State University (MSU), the Academia and members of press for gracing the occasion and making it a reality. Dr. Munthali reminded the audience that the project will be implemented in four years period thus and was supposed to start from January 1, 2013 to December 31, He therefore requested the collaborators to consider extending the implementation period in order to redeem the lost time so as to make full use of the whole allocated time (four years period). Understanding that the project is implemented in about 12 countries in Africa Malawi inclusive, the Minister said that Malawi might have been included either:...because it has a good reputation abroad or it has a lot of challenges which the project is geared to address. On a good note Dr Munthali realised that the project Guiding Investments in Sustainable Agricultural Intensification in Africa is in tandem with the Agriculture sector investment framework and will help the government to attain the Agricultural Sector Wide Approach (ASWAp), ultimately achieving government`s goal of ensuring food security at both household and national levels. On the objectives of the project particularly objective 1 devising strategies to promote the sustained, profitable use of fertilizer and improved seed for the main food staples, MAIZE in particular Dr Munthali questioned the mention of maize reminding the audience that the historical staple crop is no longer a viable option for Malawi due to climate change that has affected most districts of the country. He argued that Malawi has now to start thinking of diversifying away from maize to other equally promising cereals such as sorghum and Millet in areas like Balaka, Nsanje, Chikwawa etc. He added that the objectives are such framed that the careful implementation might lead to a green revolution in Malawi and pledged undivided government support in the project where need be. On the argument that an estimated investment of $ 2.5 billion government expenditure on subsidy in Sub-Saharan Africa (SSA), effecting an improvement in effectiveness in these programmes by 1 per cent yielding up to $ 250 million savings, Dr Munthali Centre for Agric. Res. & Dev, Malawi 1

8 said that we seem not to be too ambitious because if say we aim at 5 per cent improvement in effectiveness, the governments will automatically save huge amount of dollars ($1.25 billion) for alternative uses. Further than that, FISP expenditure (about 61 billion) for Malawi is a huge investment hence devising ways to improve the effectiveness of the system is a welcome development and promised that government will follow through the proceeding of the project up to the end. The Minister said that government will be eagerly waiting for another workshop of the same kind to share notes on the research findings and commended that collaboration in running the project assures sustainability. DR Munthali then applauded the collaborators and all participants for availing themselves for the workshop despite their tight schedules. At exactly 10:10am the Minister of Agriculture and Food Security Dr James T. Munthali officially declared the workshop open for further proceedings. Dr Kenneth Wiyo, the Dean of the Faculty of Agriculture at LUANAR who represented the vice chancellor, welcomed the guest of honour and all participants to the event. He noted that FISP being chronic in its implementation might meet numerous challenges demanding the researcher s address. He called for active and impact-oriented contributions that can revive the green revolution in our country. 3.0 WORKSHOP OBJECTIVES 3.1 Broad objective of the workshop The seminar was organised with the primary objective of soliciting views from stakeholders on how to make FISP more profitable and sustainable for smallholders by bringing new insights on the design of the proposals to be carried out by CARD staff. 3.2 Specific objectives of the workshop To help in the design of research proposals to be carried out by LUANAR that will bring new insights on improving the efficiency and profitability of FISP to farmers. To provide recommendations on how to make FISP more sustainable in the face of financial resource constraints by government. Solicit views and comments that will contribute to shaping the project design to ensure that its objectives as well as its goal of improving incomes and food security status of farmers and consumers through contributing to sustainable agriculture productivity growth in Africa is met. 4.0 SUMMARIES OF PRESENTATIONS About seven papers were presented at the workshop which are: 1) The role of FISP in the Malawi Economy: Any Prospects of Graduation? 2) What are the farm-level impacts of Malawi s farm input subsidy program? A critical review; 3) An overview of the Guiding Investing in sustainable Agricultural Intensification in Malawi; 4) Role of Private Input Suppliers in Smallholder Agriculture: Insights from Malawi; 5) Improving Maize Productivity through the Promotion of Sustainable and Profitable Use of Fertilizer: The Role of Complementary and Restorative Management Practices 6) Profitability of Fertilizer Use: Evidence from Malawi 7) Farm Household Efficiency and Farm Input Subsidy: Evidence from Malawi Centre for Agric. Res. & Dev, Malawi 2

9 4.1 The role of FISP in the Malawi Economy: Any Prospects of Graduation? By Prof. E. Chirwa This paper was presented by Ephraim Chirwa, a Professor of Economics at Chancellor College, one of the constituent Colleges of the University of Malawi. Prof. Chirwa highlighted that 2013 is the 9th year since the start of the FISP program Whose main goal is to increase household food security of Malawian smallholder farmers through the provision of inputs at affordable prices and increasing income earned by households from food production. He admitted that it is difficult to evaluate the impact of FISP realizing that almost all the people in Malawi are benefiting from the program be it directly by receiving coupons or indirectly by either buying maize at a lower price because of low cost of production attributed to FISP or vulnerable local farmers selling cheap fertilizers to non-poor households. He noted in his presentation that Malawi would have been officially importing maize if it were not for FISP in the past few years which he said in a way we are saving forex (for reduction in maize importation) which is being used for importing other necessities through still import inputs for production. On a sad note he noted that when prices of fertilizers are going up, the government seems to be subsidizing more building in dependency syndrome instead of using the price as a good signal to the citizen by likewise raising the prices of subsidized fertilizers. He also argued that market based instruments be used to help commercial farmers not FISP for cash crops. Among other challenges, it was noted that subsidy in Malawi has failed to stabilise maize market prices which has been volatile over a long period even during the periods of bumper harvest. Not only is the subsidy having negative impacts on the private sector, but also it has been observed that a lot of Micro and Macro Agro Dealers have been established during the times of FISP thus FISP has contributed positively the employment levels in Malawi. Prof. Chirwa noticed that Smallholder farmers who were food secured before the program for 2 month have improved to 4 month and a lot of other benefits. Prof. Chirwa raised a number of questions to the audience in his presentation; a) What if there was no subsidy would the situation be better or worse? b) Why has poverty level not fallen in FISP period? c) Is it possible to graduate from FISP On the graduation, Prof. Chirwa emphasized the need for any development program to build in aspects of sustainability overtime. He argued that the subsidy data for 2013 is indicating that about 50% of the FISP beneficiaries have received coupons since the program started indicating little or no graduation of the larger masses. He however, suggested that beneficiaries could have been told that if they receive coupons for a fixed number of times say three, they should be giving way to others knowing that subsidy only goes to a population of about 1.5 million out of 15.4 million population. Prof. Chirwa also proposed that perhaps the dilution of the benefits at local level where people share a 50kg bag of fertilizer might be one of the major causes of low productivity at household level and asked the stakeholders to pay a closer attention to the malpractice. In agreement with Dr Munthali, Prof. Chirwa seconded the Minister that maize should give way to other cereals that are already performing in Malawian context. He argued that maize might not be the staple food for our ancestors 100 years ago just as sweet potato is a staple food for Europe why should we continue producing a crop that is giving us trouble all the time he posed a question to the audience. 4.2 What are the farm-level impacts of Malawi s farm input subsidy program? A critical review by Dr. Rodney Lunduka, IIED, United Kingdom The paper was presented by Dr. Rodney Lunduka from the Institute for Environment and Development in the United Kingdom. Dr Lunduka started by posing a question to the participants on their perception of the goal of the FISP: The question was who thinks the goal of FISP is: a. Poverty reduction b. Increasing maize production c. Both d. Neither None of the participants choose option (a), 15 participants supported (b), 14 participants opted for (c) and the other 14 were in support of (d) which is telling us the that different stakeholders vary in their understanding of the program`s objectives. Dr Lunduka said that his presentation was a critical review of papers that researchers have published about Malawi`s FISP more particularly up to the year 2011 and that this will help government and stakeholders to identify policy gaps and avoid reinventing the wheel on the side of researchers. In his analysis Dr Lunduka noticed that there is more cost than the benefits realised by farmers due to low productivity. Among other challenges the study reviewed that FISP there is 1) high politicization (identifying suppliers, procurement, even the actual identification of beneficiaries) Centre for Agric. Res. & Dev, Malawi 3

10 2) late administration of coupons leading to late application consequently poor yields for smallholder farmers 3) Misunderstanding of the Government figures which are indicating an increase in maize production, but even though the production of maize is going up, the prices for maize are increasing also. This is contrary to the economic theory of demand and supply of basic economic theory and thus poses a question on what is really happening on the ground. Hence commended a study to be conducted to screen out what is really happening. Issues he raised in his review a. Targeting a) Rich farmers are more productive than poor and the poor have no managerial skills at least be providing labour to the efficient farmers b) How to implement FISP without hampering the performance of the private sector which is integral to the performance of the program c) Is it possible to reduce poverty by giving out cheap fertilizer to the poor? He finished by proposing that there be an Increase the private sector participation in sourcing, transportation and final retailing of fertiliser Improve on the time for delivering of inputs Introduction of extra programs that would be aimed at the very poor households i.e. social transfer programs and safety Nets. 4.3 An overview of the Guiding Investments in Sustainable Agricultural Intensification in Malawi by Jacob Ricker-Gilbert, Purdue University Dr Jacob presented on the GISAIA project proposal and invited the private sector, development partners, the academia and all partners to help in both the design and implementation of the project so as to make sure that the project generates new information that will be relevant to Malawians. He said that the virtue that the project is implemented in 12 Africa countries, cross border reviews will help countries to learn from each other. He emphasised the point that the aim of the workshop was to solicit views from the audience that would help the project clearly and promptly address the issues of FISP. 4.4 Role of Private Input Suppliers in Smallholder Agriculture: Insights from Malawi by Charles Jumbe, CARD Dr Jumbe presented the research proposal developed by Mr. Stevier Kaiyatsa who was in Pretoria, Republic of South Africa (RSA) taking some courses for Masters Degree. The proposal highlighted that input supplier s remains inseparable to the success of the Farm Input Subsidy Programme in Malawi. 4.5 Improving Maize Productivity through the Promotion of Sustainable and Profitable Use of Fertilizer: The Role of Complementary and Restorative Management Practices by Jonathan Kanyamuka, Bunda College, LUANAR Mr Kanyamuka highlighted that fertilizer on its own might not be a lasting and sustainable solution for farmers in Malawi hence other restorative and complementary measures such as intercropping, timely planting, use of compost manure, farm yard manure, conservation Agriculture among others which was in agreements with the two background papers presented. 4.6 Profitability of Fertilizer Use: Evidence from Malawi, Mr. Francis Darko, Purdue University In his presentation, Mr Darko, a PhD student at Purdue University noted that Malawi has low productivity as compared to its neighbouring countries of South Africa and Zambia in both cereal and legume crops. He observed that there is a room for increasing production since the yield gap ranges between 38 to 53 percent for cereals and 40 to 75 percent for legumes. 4.7 Farm Household Efficiency and Farm Input Subsidy: Evidence from Malawi, by Mr. Francis Darko, Purdue University Mr Darko observed that the changing technical and economic environment in which farmers are exposed to demands critical decision making on the allocation of resources to different enterprises. In his presentation he noted that there is a correlation between subsidy program and the amount of land and other resources allocated to various enterprises citing an example of Kasungu that farmers struggle to strike a balance between the land allocated to maize for FISP and tobacco as a cash crop. Mr Darko suggested that much as it is important to improve on the adoption of new technologies and providing off-farm income opportunities to these farming households, the gap between the actual and potential yield is another promising channel to improving the economic livelihood of farmers. Centre for Agric. Res. & Dev, Malawi 4

11 5.0 CONCLUSION AND FOLLOW UP ACTIONS 1. Did you investigate the possibility of tying the FISP to some socially desirable goal like: Following the presentations, participants commented, gave their suggestions and some of them asked questions as presented below: (a) (b) (c) Use of manure or composting Planting of LEGUMES Planting of TREES (CHIRWA) Prof. Chirwa expressed a concern over the prices of FISP fertilizer over time, saying that we have failed to make our smallholder farmers productive by giving them bad price signal suggesting that price is the best signal that the government could have implemented. On whether FISP is a social program that the government is implementing, Prof. Chirwa reminded the audience that about 58% of the rural population in Malawi is living below poverty line and that FISP is not even able to help half of the said percentage. He therefore suggested that proper targeting be instituted as weather subsidy should target the poorest of the poor or otherwise. Further than that he suggested that if we can target the marginally poor few resources can push them to the other side of the non-poor. Prof. Chirwa argued the house that a study has shown that a cash transfer program of 10 billion can push the poor out of their menses. On his part Dr Lunduka suggested that probably farmers be grouped into three groups 1) Commercial farmers who should be targeted at the output market, 2) Farmers with a landholding size of 1-2 hectares (needs to be pushed up) 3) Poorest (unproductive) farmers should be providing labour ( ganyu ) to the commercial farmers in a way productivity might be obvious. A research should be conducted to find out whether the increase in price with increase in production is a result of the movement of maize out of the country at the borders or because of the loss of the grains at the storage facilities due to rotting. Nutritional effect changes that have occurred since the implementation of the FISP be looked at as an indicator of food security. Integrated soil management practices should be included in FISP for the higher returns. The effects of FISP on other agricultural programs like extension should be researched on understanding that it constitutes a lion s share of the total allocation to the MoAFS in the National budget. 2. Although FISP has not achieved positive gains in cost-benefit, shouldn t we look beyond costbenefit in monetary terms, and factor in the ultimate end; improved nutrition? Improved nutrition has generational benefits on human productivity, intellect, reduction in diseases and such unspoken benefits that come not in the short term, but in the long term. (ALEXANDER KALIMBIRA) 3. It seems the FISP is a social programme. Also it has too many objectives! Could we target so as to address the commercial farmers and poor people to be addressed by social programmes. (I KUMWENDA) 4. Coming from a background that the core objective of FISP is to increase food security which entails increased production my question is: Can the FISP target those that have resources to use subsidized inputs in order to produce more and the marginalized can buy the commodity at reasonable price: - It saves in implementation cost - Estates or farms will employ more 5.(a) FISP is very packed to resolve a lot of issues enhancement of income, food security there is need to unpack it into separate programmes: (i) Commercial Farming Programme (ii) FISP for very poor farmers 5(b) Extension services are also very important that can enhance productivity. Targeting should embody extension services. 5(c) There is need to distinguish between the very poor and economically active persons in the design of FISP programmes. 5(d) The recent increase in maize production but with increase in prices. It could be due to the increased demand of maize more than supply. 5(e) Universal Input Supply can eliminate a lot of logistical problems. Or else use a coupon system so that farmers can buy wherever, other than having specific suppliers of subsidy fertilizer. 6. What is a critical comparative analysis of FISP and non FISP beneficiaries in profitability of subsidy in Malawi? Centre for Agric. Res. & Dev, Malawi 5

12 7. How does strategic grain reserve and cross border trade affect the maize production and price in Malawi recently as both are increasing? Are there research being done to find the linkages on maize production and prices with cross border/strategic grain reserve management? 8. It is important that existing strategies have to be found. The situation at farmer level is that farmers practice integrated soil fertility technologies where they combine organic soil fertility technologies with organic fertilizers. The amount of fertilizer is reduced while farmers improve their yield. The project should therefore help in testing and validating to organic technologies that can be combined with inorganic fertilizers. 9. Has FISP impacted on other mainstream budgetary allocation to the agriculture sector like extension? 10. Interesting and promising research work but what is the timeframe as policy decisions on redesigning FISP may have to come pretty shortly? 11. Small Credit Association (SACA) should read Smallholder Agriculture Credit Administration. 12. Research should go into market participation of farmers. For example, why is price going up and production is also going up? 13. The capacity building component of the project (targeting MoAFS) should be carefully implemented. Often times we have provided training that is irrelevant to the Ministry. The training topics should be informed by needs identified by the Ministry. 14. What is the project doing about establishing a platform to ensure that past research, on-going research by other research institutes, projects either public or privately funded in the same themes are being taken on board in their project? Is there possibility for establishing a platform of researchers working on the same area so that policy advice is consolidated thus ensuring appropriate government policy making and programme implementation? (MARIAM MAPILA) 15. Design, Test and Validate FISP Graduation Models with various case studies. (DAVID MKWAMBISI) The project should also look at future sustainability in terms of productivity of the smallholder farmers. It should be noted that most of the beneficiaries have limited land holdings. How does this affect productivity bearing in mind the limited land holding? What other beneficiaries could be targeted by FISP in the future? How would extension services benefit subsidy programme? 16. Want to find out what the impact of adding legumes on FISP package has been. (HENRY KHONYONGWA) 17(a) Consider timing of when you collect data from private sector because some of these are seasonal and only operating in the high peak season. 17(b) The issue of crossed business is a trick issue when the suppliers are mobile especially in the seed sector. 18(a) Low production is partly done due to: - Production gaps - Low conversion rate What can we do to erase or improve this? 18(b) What is the status of agriculture extension? What measures are to be in place to improve use of technology? 18(c) - Credit How can we reduce dependency on; - Subsidy to ensure sustainability? 19. How do we reduce political element in the FISP programme as that adds to inefficiencies in the programme? 20(a) On the choice of the 4-ecological zones for the study, is it really worthwhile to include Shire Valley in the study, in view of the speech by the Hon Minister of Agriculture on propositions to focus, for example, sorghum instead of maize for Shire Valley, etc. and realization that 21(a) Sorghum is most suitable for crops for Shire Valley? 21(b) Has there been any study to find out the impact of the quality of the inputs on the FISP? Does this not have an impact on the low benefits that have been experienced versus the costs incurred? (FRED SIKWESE) 22. His work seems to be replicating the work of the maize productivity taskforces which were at Chitedze Agricultural Research Station. His study would add more value if built on flat work and not replicate it. I would rather see more of a study that assesses why those technologies were not taken up? And for those that were taken on board what has been its impact on farmer livelihood? (MARIAM MAPILA) Centre for Agric. Res. & Dev, Malawi 6

13 23(a) Profitability of fertilizer use should be considered on the pilferage that goes to other crops such as tobacco particularly on NPK. Yet the analysis only goes to maize. 23(b) The impact on the role of private input suppliers in smallholder farmers in terms of infrastructure considering that the awarding of supply contracts goes to some suppliers their main time of business is not input supplying. We have examples of contracts given to courier and electronic suppliers. Need to look at transparency and identification and awarding of contracts. 24(a) There is need to get a database from Ministry responsible for awarding contracts for a period of time. 24(b) Visit a sample of fertilizer suppliers and determine the characteristics of private sector input supplier Official Closing In his closing remarks, Mr Alex Namaona, the Director of Planning Services in the Ministry of Agriculture and Food Security observed that a lot of issues surrounding FISP to promote food production and address food security in the country. He noted that the discussions that followed the various presentations are crucial and at the same time sensitive given the political importance of this program. He called on the researchers to design the research that will address the observed challenges in the implementation of the program and improve the design of the program to achieve program effectiveness. Finally, he thanked the Bill Gates Foundation for the unconditional support rendered to CARD to undertake this research. He further encouraged Purdue and Michigan State University to continue collaborating with CARD to address pressing problems affecting the agricultural sector. 24(c) Make a recommendation on increased participation of fertilizer input suppliers from private sector. At the end of the workshop it was observed that: Targeting the poor to increase production might not necessarily be the right option rather productive farmers. Poor smallholder farmers which constitute a large population of the country be targeted in other government programs such as safety nets and cash transfers. A study should be commissioned to find out why increase in maize production is accompanied with an increase in prices against basic theory of demand and supply. The cost for FISP is higher than the benefits (Yield) The Director of Planning Services in MoAFS Mr Namaona responding to the question as to whether diversifying away from maize is a policy at incubation stage, Mr Namaona said that it s a policy that the MoAFS in moving since most of the policies begins at a podium. Mr Namaona also took time to explain that contract awards for the supply of inputs by government follows a normal process and that every firm that bids is equally considered without any bribe. Centre for Agric. Res. & Dev, Malawi 7

14 Annex 1.1 Background Paper: The role of FISP in the Malawi Economy: Any Prospects of Graduation? By Ephraim W. Chirwa, Mirriam M. Matita, Peter M. Mvula and Andrew Dorward August 2013 Abstract: This paper reviews the role of the Farm Input Subsidy Programme (FISP) in the Malawian economy, and given the cost associated with its implementation whether there are prospects of sustainable graduation. The paper suggests that the impacts of the input subsidies can directly and indirectly affect macroeconomic aggregates and household welfare. These impacts are important pointers to the prospects of sustainable graduation from input subsidies. The costs of the programme have been increasing and the FISP continues to constitute a significant proportion of the budget. However, the evidence on the impacts are mixed, with growth and real wages in the rural areas improving, maize prices have remained volatile and public debt and budget deficit have been worsening, and national poverty levels have not changed substantially. The direct beneficiary impact at household level are more muted and are likely to be dominated by the positive economy-wide effects, although school enrolment and child health show improvements. These results provide more mixed prospects of sustainable graduation from farm input subsidies. Nonetheless, the analysis highlights the challenges of targeting and sharing of subsidy among households, which may have implications on the direct beneficiary impacts and prospects to sustainably graduate from the programme. 1.0 Introduction This paper analyses the impact of Malawi s Farm Input Subsidy Programme (FISP, previously known as the Malawi Agricultural Input Subsidy Programme, MAISP) on the economy and selected indicators of household welfare. The 2012/13 season marked its eighth year of implementation and some households have had continuous access while others have had intermittent access to subsidized fertilizers. Although the main objective of the input subsidy is to increase productivity and food security, it plays multiple roles and has the potential to influence other social and economic indicators of well-being. The FISP attempts to resolve the low maize productivity trap whereby large inter year maize price instability means that fear of high maize prices forces large numbers of poor, maize deficit farmers to grow as much maize as they can, even though they cannot afford to purchase high yielding seeds and fertiliser, with consequent low land and labour productivity and incomes. Substantial input price reductions through the FISP provide a means for addressing problems of both profitability and affordability, with different impacts on different types of households. This should lead to increases in labour, land and capital productivity among households. Understanding these different impacts, and how they impact on wider nonagricultural incomes and markets, is important for assessing potential processes for graduation from agricultural input subsidies. It is, however, worth noting that the issue of graduation is not part of the design of the programme, but it is an issue that some stakeholders have been raising over time. SOAS et al (2008) and Dorward et al (2010) suggest various pathways through which a large-scale farm input subsidy programme affects different types of households, different markets and the economy. These effects are classified into effects on the macroeconomic environment (fiscal, monetary, growth and food price effects), effects on input markets (displacement and investments in input supply systems) and rural household impacts (direct beneficiary effects and rural economy-wide effects). SOAS et al (2008) present a framework for understanding the different direct and indirect impacts of input subsidies on different households in a rural economy, as presented in Figure 1. The effects on recipient households arise from the direct beneficiary impacts of the subsidy programme through increased production and incomes from sales of agricultural output, resale of coupons by poor households and displacement use by less poor households. Centre for Agric. Res. & Dev, Malawi 8

15 Figure 1: Understanding Household and Local Economy Impacts of Input Subsidies RURAL HOUSEHOLDS Poorer households Resale Farm/ non farm investment Y1 Increased Y2 Increased real incomes real incomes Input Subsidy Incremental use Displacement use Less- poor households Y1 Increased production Y2 Increased production Y1 Increased wages Y2 Reduced maize prices Y2 Increased wages RURAL Input service demand Farm/ non farm ECONOMY & investment demand & investment Note: dotted lines represent negative effects for less poor maize surplus households Source: SOAS et al (2008) The other effects arise from economy-wide impacts owing to the scale of the programme through the price effects reduced price of food and increase in wages. These economy-wide effects affect both recipients and non-recipient households in the rural economy. The economy-wide impacts can also affect the macroeconomic environment and promote economic growth. The increased incomes arising from direct beneficiary impacts and economy-wide impacts may stimulate further investments and diversification in farm and non-farm activities, with implications on the overall growth of the economy. These various effects of the farm input subsidy programme depend on the implementation efficiency and the cost-effectiveness of the programme and the various shocks and stress that households experience. At household level, the size of the benefits or subsidy package, the targeting of beneficiaries, the timing of access to the subsidy and access to extension services are critical in realising direct beneficiary benefits from the subsidy programme. SOAS et al (2008) and Dorward et al (2010) highlight specific issues that can affect the direct beneficiary impacts of the subsidy such as targeting (with the better off more likely to receive the subsidy), size of the benefits (with widespread redistribution of coupons within the village), improvements in the timing of receipts and limited access to extension advice on fertilizer and seed variety use. The input market and economywide impacts also depend on the efficiency and cost effectiveness of the subsidy programme including scale of the programme, procurement, targeting and distribution of inputs. For instance, reduced maize prices and increased wage rates may kick-start growth in the rural economy while poor targeting may lead to displacement of commercial sales of farm inputs and exclusion of the private sector in the implementation of the subsidy programme may reduce private investment in input supply systems. SOAS et al (2008) and Ricker-Gilbert et al (2010) find evidence of displacement of commercial sales of fertilizers due to the subsidy programme. This paper is organized into five sections. In the next section, we review the scale and financing of the farm input subsidy programme. Section 3 presents the evidence on the impacts of subsidies on various indicators, including indicators of the economy-wide, input markets and direct beneficiary household effects of the subsidy programme. Section 4 focuses on a review of the concept of graduation and whether there is evidence of prospects of sustainable graduation. Finally, we offer concluding remarks. 2.0 Scale and Financing of the FISP 2.1 Principal Features and Scale of FISP 2005/6 2011/12 The core stated objective of the FISP has consistently been to improve resource-poor smallholder famers access to improved agricultural inputs in order to achieve their and national food self-sufficiency and to raise these famers incomes through increased food and cash crop production. Later years of the programme have given greater emphasis to concerns for vulnerable farm households. Throughout this, however, there has been an emphasis on programme beneficiaries as farmers and producers, with little emphasis on beneficiaries as consumers and there has Centre for Agric. Res. & Dev, Malawi 9

16 been no emphasis on the indirect programme effects on maize prices and hence consumers. Table 1 presents the principal features of the programme since2005/06 with fertilizer subsidies increasing from 137,000 metric tons in 2005/06 to 170,000 MT in 2007/8 and falling to 140,000 MT in 2011/12. In most of the years, the actual fertilizer subsidized has tended to exceed the planned supply, with the exception of 2005/6, 2009/10 and 2011/12 season. Another interesting trend is the increase in the level of the subsidy from MK1,750 per 50 kg bag in 2005/6 to MK6,536 in 2011/12, with a resulting Table 1: Principal Programme Features, 2005/6 to 2011/12 reduction (nearly constant) redemption price (farmers contribution). Maize seed subsidization started in 2006/7 to complement the subsidy on fertilizers while in 2007/8, subsidization of legume seeds started to address the nutrition aspects in the programme. The earlier period of the programme also covered the subsidization of fertilizers for tobacco, which was abandoned in 2009/10 agricultural season. Fertilizer subsidies on smallholder tea and coffee were only implemented in 2008/9 agricultural season while subsidies on cotton seeds and chemicals were implemented in 2007/8, 2008/9 and 2011/12 agricultural seasons. 2005/6 2006/7 2007/8 2008/9 2009/ / /12 Fertiliser voucher distribution (MT equivalent) 166, , , , , , ,000 Total subsidised fertiliser Planned 137, , , , , , ,000 sales (MT) Actual 131, , , , , , ,901 Fertiliser voucher value, approx. (MK/bag) 1,750 2,480 3,299 7,951 3,841 5, Redemption price (MK/bag) 950* Subsidy % (approx.) 64% 72% 79% 91% 88% 91% 93% Subsidised maize seed (MT) n/a 4,524 5,541 5,365 8,652 10,650 8,245 % Hybrid seed 0% 61% 53% 84% 88% 80% 68% Legume seed (MT) ,551 2,726 2,562 Cotton seed / chemicals No No Yes Yes No No Yes Total programme cost planned 5,100 7,500 11,500 19,480 21,908 19,700 21,586 (MK million) actual 4,480 10,346 13,362 33,922 15,526 21,868 23,455 Note: * 950MK per bag for maize fertilisers and 1,450MK per bag for tobacco fertilisers Source: Chirwa and Dorward (forthcoming) 2.2 Financing and Costs The FISP has largely been financed from domestic sources and indirectly through budget support from the country s multilateral and bilateral donors. The seed component of the programme has, however, been directly funded by donors. As Chirwa and Dorward (forthcoming) note, there are considerable difficulties in obtaining the true costs of the subsidy programme in Malawi due to poor accountability of farmers contributions from ADMARC, incidence of carryover stocks, uncertainty about the human resource costs of implementing the programme (stafftime and the ORT in various departments involved in the implementation). Figure 2 shows the trend in the costs and costs components of the FISIP since 2005/06 agricultural season. Programme costs appeared to be increasing exponentially from 2005/6 to 2008/9, while the programme budget was supposed to rise more slowly and steadily. This led to an increase in estimated expenditure from US$74 to US$250 million from 2006/7 to 2008/9, with the programme accounting for 68% and 16% of the MoAFS and national budgets and 6.6% of GDP in 2008/9. However, while the 2009/10 budget continued to rise, 2009/10 actual programme costs fell dramatically below the budget, and although costs rose again above the budget in 2010/11 and 2011/12 they remained below half of 2008/9 levels in relation to the national budget and GDP (at or below 8% and 3% respectively). The major cost item causing programme cost changes is clearly the cost of fertiliser (which includes procurement and transport to markets), as seed and other costs are responsible for a small proportion of programme costs and show a steady increase over the life of the programme (due to increases in seed volumes and prices and greater reporting of other costs items). Centre for Agric. Res. & Dev, Malawi 10

17 Figure 2: Budgeted and Actual Programme Costs % 60% 50% 40% 30% 20% 10% 0% 2005/6 2006/7 2007/8 2008/9 2009/ / /12 Other costs (US$ million) Seed costs (US$ million) Fertiliser costs (net) (US$ million) Programme budget (US$ million) Cost, % MoAFS budget Cost, % national budget Cost, % GDP Source: Chirwa and Dorward (forthcoming) Fertiliser costs are then the product of volumes and prices, and fertiliser volumes increased dramatically from 2005/6 to 2007/8, and then fell back somewhat in 2008/9 (though they were still above 2006/7 volumes) before being rigorously cut and held to budget in from 2009/10 onwards. The major cause of the price spike in 2008/9 was high international fertiliser prices. The pattern of rising fertiliser volumes from 2005/6 to 2008/9, the commitment to these volumes despite very high prices and costs in 2008/9 and then the rigorous cut in volumes after 2008/9 have to be seen in the context of the changing political circumstances before and after the May 2009 Presidential and Parliamentary elections. 3.0 Impacts of the Farm Input Subsidies The analysis of the impact of the subsidy programme is categorized into three: macroeconomic effects, maize price and wages effects, input market effects and direct beneficiary household effects. 3.1 Macroeconomic Effects The macroeconomic environment since the introduction of the farm input subsidy programme has remained relatively stable. Table 2 shows trends in some of the macroeconomic indicators between 2005 and From 2005 up to 2008 the economy witnessed increases in the both agricultural and gross domestic product. Since 2009, the economy has witnessed a decline in the growth rate but it has still been growing at above 6 percent. Agricultural output grew by 6.6 percent in 2010 compared to 10.4 percent in The reduction in agricultural growth rates have been attributed to the dry spell that hit some parts of the country. The overall growth rate in gross domestic product in 2010 was largely helped by the 53 percent growth rate in the mining sector, implying that growth could have been much lower without the emerging mining sector. Nonetheless, both the growth rates in gross domestic product and agricultural output have been partly attributed to the subsidy programme and the good rains that the country has witnessed over the past 6 agricultural season. Table 2: Macroeconomic Performance Indicators, (%) Indicator Real Agricultural Growth Real GDP Growth Inflation Deficit/GDP Ratio (actual) Deficit/GDP Ratio (budget) Debt/GDP ratio Source: Reserve Bank of Malawi, Financial and Economic Review, 22 (4), 2010 The deficit/gdp ratio in the fiscal budget has been worsening, particularly up to 2009 from -1.5 percent in 2006 to -8.2 percent in However, based on projected actual figures, there was Centre for Agric. Res. & Dev, Malawi 11

18 expectation of a surplus of 1.6 percent of gross domestic product in More worrying is the increase in the indebtedness of the country from 8.2 percent of gross domestic product in 2006 to 15.7 percent of gross domestic product in The peak in domestic debt appears in 2008/2009, which also witnessed high fiscal deficit/gdp ratio and this was also the year the subsidy cost was 6.6 percent of gross domestic product and the subsidy budget was over-spent by about 87 percent, partly due to higher fertilizer prices and partly due to expansion of the programme (Dorward et al, 2010). There has also been price stability over the period of implementation of the farm input subsidy, with inflation on a declining trend from 15.4 percent in 2005 to a single digit level of 7.4 percent in 2010, although maize prices rose dramatically from early 2008 to early 2009, before falling back in mid-2009 to Figure 3, right panel, shows that inflation continued to fall owing to the low prices of maize that have been experienced in Maize prices account for a significant proportion of the food component of the consumer price index, and reduction in maize prices have exerted downward pressure on the general price level and food inflation. Reductions in the price of maize in 2006/7 and 2009/10 are attributed to the economywide effect of the subsidy programme that improved availability of maize in the economy 1. These positive macroeconomic developments have also been accompanied by reduction in the projected incidence of poverty as shown in figure 2, left panel. Since 2006, the poverty rate based on the model-based prediction fell from 52 percent to 39 percent in It is not clear why maize prices rose in 2008/9 and without apparent hardship for the poor probably due to a combination of rising ganyu wage rates and disruption of a thin market by official export of over 300,000MT of maize in late 2007 when it was thought that maize stocks were higher than they actually were (Dorward and Chirwa, 2011). Centre for Agric. Res. & Dev, Malawi 12

19 Figure 3 : GDP Growth, Agricultural Growth, Poverty and Inflation, Percent Percent Source: Computed by authors based on data from Reserve Bank of Malawi and NSO data 3.2 Maize Prices and Wages One of the expected benefits of the farm input subsidy is to reduce both the price of maize relative to rural incomes. Between 2001 and 2011, there are three surges in prices of maize: in 2001/2, 2005/6 and 2008/9. Price surges in 2001/2 and 2005/6 are mainly explained by reductions in maize production owing to poor weather conditions (with heavy rains in March and dry spells and floods in some areas exacerbated by low input uptake in 2000/1and late distribution of inputs and poor rains in many areas in 2004/5) (Chirwa, 2009). The surge in maize prices in the 2008/9 market season should not be attributable to such supply side issues because of relatively good rains and improved access to subsidised seeds and fertilizer. High prices in 2008/9 (and other market seasons) also raise questions about MoAFS high national maize production estimates. Figure 4 shows 1993/4 to 2010/11 maize prices (average annual prices from MoAFS market surveys, in US$) against estimated quantity consumed per capita, calculated from Ministry of Agriculture and Food Security crop production estimates, census data, and exports and import estimates compiled from various sources.this perspective on prices and supply estimates draws attention to an apparent shift in the relationship between net supply and prices from around the 2006/7 production season. \ Centre for Agric. Res. & Dev, Malawi 13

20 Figure 4 : Maize prices and Estimated Quantity Consumed Per Capita, Mean maize price 2005 US$/kg /1 1993/4 1994/5 2004/5 2007/8 2006/7 2010/11 production seasons 1997/8 2008/9 1996/7 2001/ /9 2010/ /4 to 2005/6 1995/6 production seasons 2009/ / /4 2002/3 2005/6 1999/ Per capita maize crop estimates plus net imports (MT) 1995/6-2004/5 2005/6-2009/10 Regression estimate, Ed=-0.51 Source: Chirwa and Dorward (forthcoming) Chirwa and Dorward (forthcoming) also find that monthly variability of maize prices (measured in standard deviations) has also increased in the FISP period, although the coefficient of variation has fallen slightly. Prices and price variability have also increased in US$ terms, though not as dramatically. It appears therefore that the subsidy programme has not significantly reduced either prices or food price risks. Nonetheless, this does not necessarily mean that the FISP has not exerted downward pressure on maize prices but rather, that other pressures pushed prices up in 2008/9. These pressures include export market speculation, ADMARC and NFRA purchasing behaviour, high international maize prices and Government s stockpiling to feed into newly constructed strategic reserves (Chirwa, 2009). In the absence of such pressures maize prices did fall and remain low throughout the 2006/7 marketing season, following the introduction of FISP in 2005/6 and before government interventions tightened the market after the 2007 harvest. Prices were also low throughout the 2010/11 market season, prior to the macroeconomic problems that surfaced in mid One of the expected economy wide benefits of such a large scale input subsidy programme is its influence on rural wages relative to maize prices. The wage effects of input subsidies can be gleaned from household survey and qualitative interviews data. Figure 5 shows the levels of maize and tobacco prices and ganyu (casual labour) wage rates between 2009 and With respect to maize prices, overall the prices at which households bought maize was below Malawi Kwacha (MKW) 30 per kilogram 2, except for January 2010 (figure 5a). Generally, Blantyre and Thyolo experienced higher maize prices while Lilongwe and Kasungu experienced lower maize prices. Tobacco prices generally fell between 2009 and 2010 (figure 5b), although in Blantyre and Zomba households reported improved tobacco prices compared to the previous season. With respect to wages (figure 5c), there is an increase in wages over time as reported by households, and these increases occurred in all the districts surveyed. In terms of levels, in Mzimba and Kasungu households reported the highest wage rates while Thyolo and Phalombe households reported the lowest wage rates. These wage rates and maize price developments were also widely reported in focus group discussions and life histories of some of the beneficiaries. In most beneficiary life histories, among poor households, engaging in ganyu to earn income to purchase food is a common strategy and such improvements in wages and reduction in maize prices made maize more affordable even for poor households. This is confirmed in figure 3d which shows real increases in ganyu wages in terms of its maize grain purchasing power. Overall, the maize purchasing power of daily ganyu wages increased by 47 percent between 2 The average exchange rate in 2010 was MK150=1 US Dollar. Centre for Agric. Res. & Dev, Malawi 14

21 January 2009 and January 2010, with the highest increase of 80 percent in Ntcheu and lowest increase of 34 percent in Phalombe. Since these increases in real ganyu rates benefit both recipient and nonrecipient households, the results suggest that the rural economy-wide benefits of the subsidy programme are very strong. These high wages also enabled poor households to spend less time on ganyu in order to earn income adequate to purchase food whenever their own stock runs out. This reduction in time spent on ganyu was universally reported in focus group discussions and life histories of beneficiary households. For earlier years of the FISP, survey and FGD work in 2006/07 demonstrated similar processes of falling maize prices, rising wage rates, and falling time spent on ganyu from 2005 to Surveys and FGDs in 2009 suggested that from 2007 to 2009, rising maize prices and constant nominal ganyu rates led to some fall back in real ganyu rates. This has then been strongly reversed from 2009 onwards. Figure 5: Average Maize Prices, Tobacco Prices and Ganyu Wages a) Maize b) Tobacco MKW/da MKW/Kg MKW/Kg c) Ganyu d) Real KG of Source: Chirwa and Dorward (forthcoming) 3.3 Impacts on Farm Input Markets The farm input subsidy programme can have several impacts on the input market system depending on the scale, targeting and other implementation modalities. On one hand, a poorly targeted large scale programme results in displacement of commercial sales and introduces disincentives for private investments in input markets. On the other hand, well targeted programme can stimulate additional demand for commercial fertilizers among subsidized households by improving the productivity and profitability of their farming activities and their ability to finance fertiliser purchases. Table 3 shows the quantity of subsidized and commercial fertilizers acquired by households in 2009/10 and 2010/11 seasons by IHS2 poverty status compared with commercial fertilizers in the IHS2. Among poor households the average quantity of subsidized fertilizers declined from 54 kilograms in 2009/10 to 47 kilograms in 2010/11 while commercial fertilizers increased from 48 kilograms to 61 kilograms. A similar trend is observed among non-poor households, and may be related to economy-wide Centre for Agric. Res. & Dev, Malawi 15

22 impacts of the programme. The data also show that both poor and non-poor households supplement subsidized fertilizers with commercial fertilizers, but among the poor the higher the number of seasons a household benefits from the subsidy the lower the supplementation with commercial fertilizers. No consistent pattern emerges with respect to non-poor households that are subsidized. Table 3: Table 4 Quantity of Subsidized and Commercial Fertilizers by IHS Poverty Status (kg) Poor Households in IHS2 Non-Poor Households in IHS2 Times of Subsidy Access Subsidy Commercial Subsidy Commercial N / N / All Source: computed by authors based on IHS2 and FISS3 data A comparison of the 2009 and 2010 commercial purchases with 2004/05 purchases shows a mixed picture among different households. On one hand, among the category of poor households only those that have had access to the subsidy over 1 season and 3 seasons are on average purchasing more in 2010 than in 2004/05. On the other hand, among the nonpoor households only for households that have had access to the subsidy in the past 2 and 4 seasons do we witness purchases above the 2004/05 levels. This suggests some crowding out of commercial fertilizer sales due to the subsidy programme, although the decline in commercial purchases also occurred among households that have never received subsidized fertilizers. However, it should also be noted that the average prices of commercial fertilizers have substantially increased from MK37 per kilogram in 2004/5 to MK97 per kilogram in 2010/11, an increase of 162 percent over the period; this might have dampened the demand for commercial fertilizers. Chirwa et al (2013), using a panel of households between 2004 and 2010, find that the demand for commercial fertilizers is positively associated with number of adult equivalents, years of education of household head, maize prices, initial access to credit and value of assets; and it is negatively associated with quantity of subsidized fertilizers and poverty. Their results show that a 1 percent increase in subsidized fertilizers reduces demand for commercial fertilizers by 0.39 percent. This suggests that subsidized fertilizers displace commercial fertilizer purchases among those who purchased fertilizers in seasons. These households accounted for 54.1 percent of the total subsidized fertilizers in the sample, and using the relative shares of subsidized fertilizers we obtain weighted elasticity of -0.21, as the overall effect of subsidized fertilizers on commercial demand. Furthermore, Chirwa et al (2013) used a sub-sample of households that purchased commercial fertilizers either in 2004 or/and 2010 (capturing those households that might have entered the commercial market during the subsidy period hence those that did not buy in 2004 but bought commercial fertilizers in 2010). Their results show that a 1 percent increase in subsidized fertilizers leads to a 0.29 percent reduction in the demand for commercial fertilizers among those that purchased commercial fertilizers in either 2004 or/and The weighted elasticity using relative shares of subsidized fertilizers is for the whole sample of panel households. This Centre for Agric. Res. & Dev, Malawi 16

23 elasticity is lower than the observed for panel households that initially bought commercial fertilizers in 2004, implying that the subsidy may have encouraged some demand for commercial fertilizers. Chirwa et al (2013) also find that the demand for commercial fertilizers also falls for poor households and households that participate in labour markets but increases with number of adult equivalents, education, land holding size, average maize prices and value of assets. The demand is also much higher in the central region and northern region compared with the southern region, possibly due to the cultivation of tobacco in the central and northern region. 3.4 Food Availability and Consumption National food production and per capita availability indices show increases in national food security in Malawi since the implementation of the FISP in 2005/6. Chirwa and Dorward (forthcoming) raise questions about the reliability of some of the national crop production estimates on which these are based (possible discrepancies between maize production estimates and prices were discussed earlier, and there are long standing queries about inconsistencies between national crop estimates and survey estimates of root crop areas and production ). Nevertheless, there is general agreement that the food security situation has improved in the country partly owing to incremental use of fertilizers and improved seeds provided under the subsidy programme and partly owing to the good weather conditions that Malawi has had during this period. These trends of national food security are consistent with household survey findings that show improved self-assessment of food security. However, National Statistical Office (2012) finds that 33% of households experienced situations on food insecurity, with 42% of the rural population being food insecure in 2010/11. Concerns also arise with large numbers of people continuing to experience food insecurity and need emergency humanitarian support. This appears to be particularly serious in 2012/13 with estimates that about 1.6 million people in 15 districts, mainly in southern Malawi will be unable to meet adequate basic food requirements (FEWS NET, 2012). This is mainly due to prolonged dry spells and poor rains: similar situations in previous years have been associated with local droughts or floods and are arguably unrelated to the FISP. 3.5 Incomes and Poverty According to Chirwa and Dorward (forthcoming) several studies point to the FISP impacts on real wages implicitly or explicitly linking this to improved real incomes. Estimates of subsidy impacts on nominal wages (Ricker-Gilbert, 2011) when combined with estimates on maize production also suggest increases in real incomes for poorer nonbeneficiaries as well as beneficiaries. The informal rural economy partial equilibrium model of Dorward and Chirwa (2012) also estimates substantial real income gains from wage and maize price change impacts, averaging 10% and 3% across all households in the basic scenario in the Shire Highlands and Kasungu Lilongwe Plains respectively. In this poor beneficiary households gain most, non-poor beneficiaries and poor nonbeneficiaries gain direct and indirect benefits respectively, and non-poor beneficiaries generally lose from the indirect impacts through higher wages and low maize prices. In the Shire Highland Livelihood Zone (SHI), simulated indirect gains to real incomes for target households are considerably higher than the direct gains from subsidy receipt (13% as compared with 7%). These indirect gains are higher than simulated for the Kasungu Lilongwe Plain (KAS) because of the former s high rates of poverty incidence, high land pressure and larger numbers of poor people relying more on sales of ganyu labour and spending a higher proportion of their income on maize purchases. Chirwa and Dorward (forthcoming) argues that if the farm input subsidy raises the income of the poor then it should also play a role in poverty reduction. However, the evidence on changes in poverty at the national level is mixed. The national head count poverty rate in 2005, prior to the implementation of the subsidy programme was estimated at 52% while in the rural areas it was estimated that 56% of the population were living below the poverty line. Seasonally adjusted model-based estimates from welfare monitoring surveys suggest that the poverty rate increased in 2005/6 following the poor 2004/5 crop season and subsequent food shortages, then declined sharply between 2006 and 2007 before stabilising from 2007 to 2009 (Chirwa et al., 2012). However, recent poverty estimates based on the 2010/11 integrated household survey suggest that between 2004/5 and 2010/11 the national poverty rate is much higher than predicted by the WMS estimates, and only fell by 2 % between 2004/5 and 2010/11, suggesting only a marginal change in the wellbeing of the population. This would be consistent with high maize prices putting a brake on growth in real incomes in these years: one would then expect low maize prices in 2010/11 to stimulate further growth in real incomes and falling poverty. However, poverty incidence in rural areas is estimated at over 56% in 2010/11 (National Statistical Office, 2012), much higher than expected, with a fall of only 1.5%from 58.1% in 2004/5 to 56.6% in 2010/11. Centre for Agric. Res. & Dev, Malawi 17

24 positive impact of subsidy receipt on school attendance. 3.6 Direct Beneficiary Household Impacts Household Food Security Improvements in maize production should lead to improved food availability and food security for beneficiary households. In all the panel surveys, households were asked whether they considered their food consumption in the month before the survey to be inadequate or adequate. In order to assess the impact on food security, we created a dummy variable representing adequacy in food consumption equal to one if the household revealed that food consumption was adequate or more than adequate, and to zero if it was inadequate. Chirwa et al (2013) find that among households that received subsidized fertilizers continuously (6 times) about 22% more than non-recipients reported adequate food production, with the coefficient being statistically significant at the 5% level. Increasing frequency of fertiliser use also led to increasing frequency of reported adequate food production. These results are consistent with the qualitative evidence of increased maize production reported in focus group discussions, which might have improved food consumption. Holden and Lunduka (2010a) also find that receipt of subsidised inputs increases the probability of households being net sellers rather than net buyers of maize, and that 66% and 69% of surveyed households reported improvements in household and community food security as a result of the subsidy programme (although 60% of the households in their sample were still net buyers of maize despite the subsidy programme) Impact on Education and Health The FISP evaluation studies have also found that the subsidy has had a positive impact on human capital development in terms of children s education and health. Enrolment of the primary school age group (5-13 year olds) improved significantly among households receiving farm input subsidies. Chirwa et al (2013) find a general increase in school enrolment between the two periods, a change that was universally confirmed in focus group discussions and key informant interviews. The results also show that households that received 1-2 times, 5 times and 6 times in past 6 seasons had significantly higher enrolment rates than those that did not receive subsidies. This is consistent with anecdotal reports on programme impacts, with focus group discussion reports (School of Oriental and African Studies et al., 2008; Dorward and Chirwa, 2010a), and with Holden and Lunduka (2010) who report that 65% of respondent households perceived that there was a Improvements in food availability at household level due to access to subsidized fertilizers may improve beneficiaries health in a number of ways through improved food security and nutrition from increased own production and income, and from increased ability to finance health care. Chirwa et al (2013) find that, on average, about 59% of households had ill under-5 members in 2004/5, but this fell to 49% in 2010/11. This impact was not commonly articulated in focus group discussions and key informant interviews. The econometric evidence of the impact of the subsidy programme on the health of children in beneficiary households shows that households that had access to subsidy at least 5 times were more likely to have under-5 children that had not fallen ill in the past two weeks of the survey. Overall, there is a negative relationship between access to subsidy and incidence of under-5 illness. Holden and Lunduka (2010) also explored people s perceptions of subsidy receipt on health, and report that 40% of respondents perceived that subsidy receipt improved health. Further evidence on the impacts of subsidy access on health, but not of access to FISP itself, is provided by Ward and Santos (2010), who examined the impact on stunting from access to Targeted Input Programme inputs. They found a significant reduction in stunting for each year of receipt of TIP inputs, and based on strong international evidence on the relationship between adult height and wages; discuss possible long term beneficial effects of increased adult height on earnings Subjective Poverty or Well-being, Real Income and Assets The other evidence on the role of farm input subsidies comes from household s own assessment of the changes they have experiences, changes in real income and asset holdings. With respect to subjective wellbeing assessment, based on a poverty ladder ranging from 1 representing the poorest to 6 representing the richest, Chirwa et al (2013) find that the mean self-assessment of well-being for households increased from 1.66 in 2004/05 to 2.34 in 2010/11, representing a 41% increase. However, the multivariate analysis, did not find any statistically significant difference between recipients and nonrecipients of farm input subsidies. As noted by Chirwa et al (2013), the subsidy programme may have only weak direct income effects on beneficiary households. The results are consistent with sentiments expressed in qualitative interviews in which most households report that they are not able to produce surplus maize which could be sold to earn extra cash income Centre for Agric. Res. & Dev, Malawi 18

25 (Chirwa et al, 2013). Some life histories with selected households revealed that although some have had access to subsidized fertilizers continuously they may still struggle to produce maize that takes them to the next harvest and have to rely on ganyu to earn income to purchase food. Small but insignificant positive effects are consistent with small direct improvements from subsidy receipt which may be overshadowed by wider positive changes affecting all households through indirect market effects of the subsidy and other positive changes from 2002/3 and 2003/4 to 2006/7 and subsequent years. In contrast with these insignificant results, however, Ricker- Gilbert and Jayne (2010) do find a significant increase in satisfaction with life with increased receipt of subsidised fertiliser between the presubsidy and 2008/9 surveys. The weak results on poverty impacts are consistent with the weak relationship between access to subsidy and, real incomes and asset accumulation. With respect to real incomes, Ricker-Gilbert (2011) finds no significant impacts of subsidy receipt on non-farm income or on total household income, although net value of rainy season crop production (a measure of farm income) is positively affected by subsidy receipt in the year of receipt (but not previous years), with each extra kg of fertiliser received increasing net crop income by MK174. Dorward and Chirwa (2012) in an informal rural economy modelling compare real income estimates for target households (that is poor male- and female-headed types) with and without the subsidy (with an average receipt of 75kg and 2 kg of subsidised fertiliser and hybrid maize seed respectively per household) but with constant prices (that is without any wider market equilibrium effects). Gains averaging around 7% (just under MK1,000) across poorer beneficiary households are estimated in the Shire Highlands with lower gains (around 4%, just under MK450) in the Kasungu- Lilongwe Plains, where poverty is less severe and poor households are less capital constrained and have lower returns to capital. SOAS et al. (2008) also state that increases in beneficiary incomes were reported in a number of focus group discussions in With respect to asset accumulation, Holden and Lunduka (2010) in examining the impacts of subsidies on the value of assets and on livestock ownership measured in tropical livestock units find a general build-up in the real value of assets from 2006 to 2009, but no evidence of direct impacts of subsidy receipt on asset accumulation. Hence, there is no evidence of a general increase in livestock endowments, or of direct subsidy impacts on asset accumulation. Similarly, Ricker-Gilbert (2011) report no significant impact of subsidy receipt on household livestock and durable assets for subsidy received in the survey year or in each of the previous three years Shocks and Stresses Changes in vulnerability of households to shocks and stresses are another possible indicator of the role of subsidy receipt on household welfare. Households experience a number of shocks and stresses and most of these shocks are agricultural related. Chirwa et al (2013) find a statistically significant positive relationship between experience of shocks and those households that have had access to the subsidy, and the magnitudes are higher for recipients that have had access less than 5 times. This is contrary to the expectation of a negative relationship, and a possible explanation is that there is some targeting of the subsidy to households who have experienced shocks. However, there is a negative relationship between access to subsidized fertilizers and the severity of agriculture-related shocks. Among households that were poor in 2004/5, households with access to subsidized fertilizers are less likely to have agriculture-related shocks as their most severe shock, but there is no clear trend to suggest that the higher the number of times household access subsidies the lower the number of agriculture-related shocks that households experience. Again these results may reflect more on the likelihoods of subsidy receipt by poor households affected by severe agricultureshocks than on the impacts of subsidy receipt on vulnerability to agriculture-shocks. In summary, the evidence on changes in shocks and stresses is rather mixed. Overall, the number of shocks experienced by beneficiary households has fallen significantly over time, although those with access to subsidized fertilizers continue to experience shocks and stresses. However, among beneficiary households, agriculture-related shocks are less likely to be the most severe shocks; hence the subsidy appears to have helped poor households to be cushioned or resilient against agriculture-related shocks. Table 4 shows that there is a decline from 24% to 13% in households that experienced lower crop yields due to weather or rainfall as most severe shocks between IHS2 and FISS3, respectively. Other agriculture-related shocks whose incidence declined were large falls in sale price of crops and large rise in prices of food. The relative importance of chronic and acute illnesses appears to have risen as a result of the decline in importance of severe agricultural shocks. Centre for Agric. Res. & Dev, Malawi 19

26 Table 4: Most Severe Shocks and Stresses Experienced by Households (%) Most severe shock experienced for panel households IHS2 FISS3 2004/5 2010/11 Lower crop yields due to weather/rainfall Crop diseases or crop pests Livestock died or were stolen Household business failure non-agricultural Loss of salaried employment or non-payment End of regular assistance, aid or remit Reduced ganyu opportunities/wage rates Large fall in sale prices for crops Large rise in price of food Short acute illness/accident of HH member Chronic illness, disability or ageing of HH member Birth in the household Death of HH member Marriage/other social event Increased expenditure demands Break-up of the household Impacts from Life Stories of Beneficiary Households 3 The other body of evidence on the direct role of farm input subsidies comes from qualitative interviews with beneficiary households. Life stories from selected beneficiaries provide insights on the impact of input subsidies on their well-being. While there are positive stories about the increase in food production at household level among most households that receive subsidies, the life histories illustrate the challenges of the programme in delivering direct benefits to beneficiary households. In most life histories of beneficiaries, particularly among the most vulnerable groups (female-headed, elderly-headed and child-headed households), the stories were that 3 This section draws heavily on Chirwa et al (2013). the subsidy programme has enabled them to produce a bit more food than when they had no access to the subsidy. The qualitative analysis points to the following issues: In most cases, households that report success with the subsidy programme are those that are well to do and were already purchasing commercial fertilizers before the subsidy programme. For instance, one beneficiary who has had access to the subsidy over 5 seasons was also buying coupons that enabled him to profit from tobacco cultivation, and claimed to have transformed his life from poor category to the well-to-do category. In households that reported receipt and use of 2 fertilizer coupons, such households are likely to talk positively about the extent to which the subsidy improved their food production for such Centre for Agric. Res. & Dev, Malawi 20

27 years compared to households that received less than 2 bags of subsidized fertilizer. Sharing of coupons is widespread. Most households that have participated in the subsidy programme attributed the perceived failure of the programme to change their lives significantly to the inadequate amounts of fertilizers obtained under the programme. This is particularly the case for households that never used fertilizers prior to the subsidy programme. Many life stories described how the full package of the subsidy was beginning to change their lives, only to experience drifting back into poverty due to the dilution of the subsidy as a result of the redistribution that takes place at village level. There is also a tendency for beneficiaries to spread the subsidized inputs thinly over a larger parcel of land. Even among households that receive 2 bags of subsidized fertilizers, there were sentiments that the subsidized fertilizer was not adequate for the amount of land the household has for maize cultivation. This is exacerbated by the lack of technical advice on the appropriate use of fertilizers, with most households expressing lack of access to agricultural extension services. There is widespread recognition that the subsidy has helped beneficiary households to produce a bit more maize and more importantly in reducing the purchase price of maize even the in lean months of January and February. Most of the beneficiaries interviewed, particularly those that are still not able to produce own maize to last them to the next season, consider a low purchase price of maize as one major benefit of the programme. Households that are not able to produce maize that lasts up to the next harvest tend to purchase from the market. Most poor households engage in ganyu to earn incomes to buy maize and most reported that ganyu wage rates have been increasing while maize prices have been falling and maize is locally available. This has enabled the poor to afford purchase of maize based on ganyu incomes which have also improved over time. Due to higher wages, households reported that they have reduced the amount of time they spend on ganyu and there has also been an increase in opportunities to operate off-farm income generating activities. Poor and vulnerable households such as femaleand/or elderly-headed households that received subsidised fertilizers rarely supplement these fertilizers with commercial purchases, leading to application of subsidized fertilizers on larger parcels of land. Generally, where subsidized fertilizers are supplemented by commercial fertilizers, such households were buying commercial fertilizers prior to the subsidy and/or they are better off households that are also receiving subsidies. The quantitative analysis also shows that the supplementary commercial fertilizers are much less for poor households than for non-poor households that had access to subsidized fertilizers. 4.0 Prospects for Sustainable Graduation 4 The concept of graduation in social protection programmes has generally been linked to issues of impacts, dependency, exit and sustainability. Graduation from social protection has important implications for outreach and cost effectiveness, as it allows providers to scale down their operations and reduce costs over time. Governments with tight budgets may be more willing to support social protection if access is time-bound or if there are clear prospects of a high proportion of target beneficiaries voluntarily exiting over time. There are several definitions on what constitute graduation from social transfers, generally embodying changes from livelihoods dependent on social protection to livelihoods that can continue independent of social protection. From a programme design perspective, social protection programmes can be open-ended or time bound. Open-ended programmes (such as pensions) are not designed with any expectation of graduation. Time-bound programme transfers, however, are temporary and implemented with complementary measures intended to enable a large number of households to build their capabilities to embark on independent livelihoods. Graduation is therefore viewed as the potential to embark on sustainable, independent livelihoods without social protection pursuing an independent sustainable livelihood. Graduation is thus a removal of access to the programme that does not leave current beneficiaries supported by the programme unable to pursue sustainable independent livelihoods. Such graduation can occur at multiple levels: household, area and national levels. At household level, individual households develop capabilities to step up and or step out to engage in independent and sustainable livelihoods. At area or national levels, sufficient numbers or proportions of households in the population develop capabilities for independent and sustainable livelihoods, allowing scaling down of the programme. However, there are complex and difficult challenges in defining and measuring graduation criteria, or determining the point at which social assistance can be terminated i.e. 4 The conception framework of graduation draws largely from Chirwa et al (2011). Centre for Agric. Res. & Dev, Malawi 21

28 the thresholds of assets or incomes that are necessary for graduation. Alternative approaches include the crossing of income poverty lines or the crossing of asset and income thresholds, which are likely to vary with household structures, initial conditions, socioeconomic and cultural context, and livelihood strategies and opportunities. Nonetheless, the extent to which graduation occurs in a social protection programme depends on many factors including targeting, the nature and value of transfer benefits, duration of access, and existence of complementary interventions that strengthen household capabilities. According to Chirwa et al (2011), for graduation to occur at national, area and household levels, the core requirement is that removal of access to the subsidy programme does not reduce land, labour and capital productivity in maize production. For this to occur, therefore, potential graduation conditions are required in some combination as a result of and during the implementation of the FISP. These comprise: Falls in unsubsidised input prices compared to pre-programme prices, with, for example, improved transport systems and management of implementation and distribution of inorganic fertilizers. Reduced requirements for purchase of previously subsidised inputs due to increased efficiency in use. This can be achieved, for instance, by greater use of high yielding seed, timely planting, more effective soil health management, timely weeding, more effective fertiliser application methods, and greater use of complementary organic fertilisers. Reduced requirements for purchase of previously subsidised inputs due to substitution by cheaper inputs through for example increasing use of organic fertilizers, legume cultivation and rotations. Increased working capital among poor beneficiary households for cash purchase of previously subsidised inputs for example through increased savings or income diversification. Poor beneficiary households diversification out of maize production through either transfer of land to other high value production use (diversification or stepping out of maize within agriculture) or transfer of land to other users (diversification or stepping out of agriculture into non-farm activities). Access to low cost credit by poor beneficiary households for purchase of previously subsidised inputs for example, by introducing innovative and low cost microfinance systems. So does the evidence so far provides indication that there are prospects of graduation from farm input subsidies? As noted above, graduation can occur at national, area and household levels. At a national level, the analysis shows that there have been improvements in the supply of maize in the economy owing to the farm input subsidy programme. In addition, there is evidence that the subsidy programme is stimulating demand for commercial purchases of inputs, particularly with reductions in displacement rates over time. If such trend continues, there are opportunities for reducing the scale of the subsidy. However, other national level indicators provide mixed prospects for sustainable graduation, particularly the lack of substantial reductions in poverty, the continued volatility in the price of maize, casting doubts on prospects of graduation at national level. The prospects of graduation at household level are more limited based on the existing evidence. While there are improvements in school enrolment and under-five health, most household level indicators point to the fact that direct beneficiary effects are not substantial. This may be undermined by the economy-wide benefits that would mask differences between beneficiaries and non-beneficiaries. Nonetheless, the analysis highlights two challenges that have implications on direct beneficiary impacts and prospects of graduation from the subsidy programme at household level: targeting and sharing of coupons at village level. These issued have been documented in Dorward and Chirwa (2011), but here we use panel data to illustrate the challenges of beneficiary targeting. Both targeting and size of the benefit package have implications on creating enabling environments for sustainable graduation from input subsidies. With respect to targeting, the targeting criteria in the FISP remain quite broad with the main criteria being resource poor households, and this has meant that a large proportion of households is eligible from the perspective of communities (SOAS et al, 2008). Although the FUM (2011) study suggests that all the households conformed to the criteria by the reason they gave for their being selected for the subsidy programme, a majority (60%) indicated that their being very poor justified their receipt of coupons but no information is provided on the status of households who did not receive the subsidy. Chirwa et al (2013) mote that non-poor households bought more commercial fertilizers on average than poor households although both received equal amount of subsidized fertilizers. This suggests that non-poor households could generally afford commercial fertilizers and receipt of coupons among them largely represented targeting inclusion errors. This is also confirmed by the fact that there was no clear pattern in the perceptions of respondent households on which group was more likely than others to receive coupons, between better off farmers Centre for Agric. Res. & Dev, Malawi 22

29 and poor and vulnerable households (Dorward and Chirwa, 2011). With respect to redistribution of subsidy coupons within the villages, this practice has been widely reported by beneficiaries and from previous evaluations which indicate that a sizable proportion of households receive less than 2 coupons per household. Dorward and Chirwa (2011) find that 58 percent of households that received coupons in 2010/11 received less than 2 coupons; an increase from 49 percent in 2008/09 (Dorward et al, 2010). Chirwa et al (2013) find that on average 69% of respondent household reported that redistribution of coupons took place in their village. Widespread redistribution of fertilizer coupons was also confirmed from focus group discussions and life histories of beneficiaries. Life histories of some beneficiaries also reported the problem that sharing of coupons diluted the direct benefit per household. The practice of redistribution of subsidized input undermines the direct impact of the programme on beneficiary households and is likely to reduce the effectiveness of the direct impacts of subsidy programme and undermines the potential for some households to graduate from the programme. The issues of targeting and dilution of the benefits illustrate the challenges in improving the welfare of beneficiaries and the limits to sustainable graduation from the farm input subsidy programme. Although some households have accessed subsidized fertilizers for more than six agricultural seasons, the direct impacts on socio-economic outcome variables are mixed and weak, and there is no consistent evidence that households that have always received fertilizer subsidy perform significantly better than those that have received subsidies intermittently or have received none in the eight past agricultural seasons. 5.0 Conclusions The farm input subsidy programme which has been implemented in Malawi since the 2005/06 season has benefited the economy and households in different ways. The 2012/13 season marked its eighth year of implementation and some households have had continuous access while others have had intermittent access to subsidized fertilizers. This paper set out to provide evidence on the role of the subsidy programme on the economy, maize prices and wages, input market systems and welfare of beneficiary households. We then examined whether the evidence on impacts so far demonstrate that there is potential for sustainable graduation at area and household levels. The FISP attempts to address profitability and affordability problems to resolve the low maize productivity trap facing smallholder farmers in Malawi. It is a targeted programme focusing on resource poor farmers by providing coupons that allow them to purchase farm inputs at highly subsidized prices. Since the 2005/6 agricultural season, between 131,000 MT and 216,000 MT of fertilizers every year have been subsidized. The FISP is a major commitment of resources in the fiscal budget as it is largely financed by the Government directly through programme specific costs (purchase of inputs and implementation) and indirectly through use of staff time and operational costs by other stakeholders. Donors have played a part indirectly by providing budget support and directly by funding the seed component and monitoring and evaluation. The subsidy costs are on average 50% of the Ministry of Agriculture budget, 8% of the national budget and 3% of gross domestic product. However, in absolute terms, there has been an upward trend in the cost of the subsidy programme from MK4.4 billion in 2005/06 to MK23.5 billion in 2011/12. The FISP plays several roles in the Malawi economy. As a major budget item, it is bound to affect macroeconomic variables such as growth, maize prices, the role of the private sector and household welfare. While the growths in agricultural output and gross domestic product have been impressive in the past eight years (more than 6%) and inflation was reportedly contained, the fiscal deficit and debts have worsened during the implementation of the subsidy programme. Maize price volatility has not improved and in some years maize prices increased amid estimated surplus maize production. The instability in maize prices over time might have undermined the positive effects on pro-poor growth that might have emerged with stable maize prices. Poverty rates have not fallen significantly in spite of improvements in food availability. However, there is some evidence that real wages in terms of the amount of maize they could purchase has been increasing. The displacement of commercial sales of fertilizer also tends to undermine the impacts of the programme, although there is evidence that the subsidy programme is to some extent stimulating commercial demand. The evidence on direct beneficiary household impacts is mixed. Overall, although qualitatively communities point to the many benefits of the subsidy programme on food security, yields and other indicators of well-being, the quantitative evidence that changes in welfare indicators are attributed to the direct effects of subsidy receipt is weak. These weak direct beneficiary impacts have also been found in earlier studies. For instance, Chirwa (2010) in an earlier impact evaluation also finds weak evidence of the direct impact of participation in the subsidy programme on beneficiary households food expenditure between 2004/05 and 2006/07. Similarly, Matita and Chirwa (2011), using panel data analysis, find mixed results on the direct impact of Centre for Agric. Res. & Dev, Malawi 23

30 participation in the subsidy programme in improving agricultural growth of beneficiary households between the 2004/05 and 2008/09 seasons among households in different income quintiles. Other studies also find no significant relationship between receipt of fertilizer coupons and asset accumulation, and mixed evidence on impacts on real incomes. The direct beneficiary effects are somehow masked by the stronger economy-wide effects, in which the subsidy benefits both recipients and non-recipients thereby weakening differences between the two groups. Nonetheless, households tend to benefit from the economy-wide impact of the subsidy programme through low maize prices and increased ganyu wage rates that have been experienced since the introduction of the subsidy programme. The subsidy programme seems to have stronger economy-wide effects than direct beneficiary household effects. Over the past 6 agricultural seasons of subsidy programme implementation, the prices of maize have fallen, contributing to macroeconomic stability through falling inflation from double digit to single digit figures. The decrease in the maize prices, together with reported increases in ganyu wage rates, has meant that the poor and non-poor can afford to purchase maize at reasonable prices. In fact, there has a been real increase in ganyu wage rates measured in terms of the amount of maize a daily wage could purchase between 2010 and Although graduation is not articulated in the design of the FISP, the impact analysis raises two challenges with implications for the direct beneficiary household impacts of the subsidy programme and their prospects for graduation. First, targeting of households remains problematic and a large proportion of non-poor households have access to subsidized farm inputs. This increases displacement of commercial sales and limit incremental production. However, the reduction in displacement in subsequent years suggests better prospects for input market development as one enabling condition for sustainable graduation. Secondly, the practice of redistribution of coupons at village level, which is largely driven by village-level politics, has led to dilution of the benefit package resulting in inefficient use of subsidized fertilizers among poor households that are not able to top up with commercial fertilizers, but cultivate larger parcels of land. This sharing of coupons tends to happen among poor households. These challenges of targeting, dilution of the benefit package and inefficiency in the application of inputs may also undermine prospects of graduation from the subsidy programme for most of the households. References Chirwa, E. W. (2009) Sustained Increases in Food Prices: Effect and Policies in Malawi.. Paper presented at the FAO Regional Workshop on Policies for the Effective Management of Food Price Swings in African Countries held on 2-3 April Kunduchi Hotel, Dar-es-Salaam, Tanzania. Chirwa, E. and Dorward, A. (forthcoming) Agricultural Input Subsidies: The Recent Malawi Experience, Oxford: Oxford University Press. Chirwa, E.W., Dorward, A.D. and Matita, M. (2011) Conceptualising Graduation from Agricultural Input Subsidies in Malawi, FAC Working Paper 029. Future Agricultures Consortium, Brighton, Sussex. Chirwa, E. W., Dorward, A. R. and Vigneri, M. (2012) Seasonality and Poverty: The 2004/05 Malawi Integrated Household Survey, in Sabates- Wheeler, R. and Devereux, S. (eds.) Seasonality, Rural Livelihoods and Development. Earthscan. Chirwa, E.W, Matita, M., Mvula, P. and Dorward, A. (2013) Repeated Access and Impacts of the Farm Input Subsidy Programme in Malawi: Any Prospects of Graduation? FAC Working Paper (forthcoming), Future Agricultures Consortium, Brighton: Sussex. Chirwa, T. G. (2010) Program Evaluation of Agricultural Input Subsidies in Malawi using Treatment Effects: Methods and Practicability based on Propensity Scores, MPRA Paper No Available at Dorward, A. R. and Chirwa, E. W. (2012) Informal Rural Economy Modelling of Programme Impacts, 2005/6 to 2010/11: Discussion Paper. Evaluation of the 2010/11 Farm Input Subsidy Programme, Malawi. London: School of Oriental and African Studies. Dorward, A. and Chirwa, E. (2010a) Evaluation of the 2008/09 Agricultural Input Subsidy Programme, Malawi: Notes on Regression Analysis of Maize Production. Paper prepared for Malawi Government and DFID (Malawi). Dorward, A. and Chirwa, E. (2010b) Evaluation of the 2008/09 Agricultural Input Subsidy Programme, Malawi: Report on Programme Implementation. Paper prepared for Malawi Government and DFID (Malawi). Dorward, A. and Chirwa, E. (2011) Evaluation of the 2010/11 Farm Input Subsidy Programme, Malawi: Report on Programme Implementation. Paper prepared for Malawi Government and DFID (Malawi). Centre for Agric. Res. & Dev, Malawi 24

31 Dorward, A., Chirwa, E. and Slater, R. (2010) Evaluation of the 2008/09 Agricultural Input Subsidy Programme, Malawi: Report on Programme Implementation. Paper prepared for Malawi Government and DFID (Malawi). Farmers Union of Malawi (FUM) (2011) Promoting the Participation of Civil Society in the Management of the Farm Input Subsidy Programme (FISP), Lilongwe: Farmers Union of Malawi. FEWS NET (2012) Malawi Food Security Outlook August 2012 to January Lilongwe: FEWS NET. Holden, S. and Lunduka, R. (2010) Impacts of the Fertilizer Subsidy Programme in Malawi: Targeting, Household Perceptions and Preferences. Noragric Report. As. Imperial College of London (ICL), Wadonda Consult, Michigan State University and Overseas Development Institute (2007) Evaluation of the 2006/7 Agricultural Input Supply Program, Malawi. Interim Report prepared for Malawi Government and DFID (Malawi). School of Oriental and African Studies (SOAS), Wadonda Consult, Michigan State University and Overseas Development Institute (2008) Evaluation of the 2006/7 Agricultural Input Supply Program, Malawi. Final Report prepared for Malawi Government and DFID (Malawi). Ward, M. and Santos, P. (2010) Looking Beyond the Plot: The Nutritional Impact of Fertilizer Policy. Selected Paper prepared for presentation at the Agricultural & Applied Economics Association 2010 AAEA, CAES & WAEA Joint Annual Meeting Denver, Colorado, July 25-27, Woodridge, J. M. (2000) Introductory Econometrics: A Modern Approach, Cincinnati: South-Western. WFP (World Food Programme) (2008) Food Consumption Analysis: Calculating and Use of the Food Consumption Score in Food Security Analysis, Rome: WFP Vulnerability Analysis and Mapping Branch. Matita, M. M. and Chirwa, E. W. (2011) Agricultural Growth and Poverty in Rural Malawi, Research Report submitted to the African Economic Research Consortium, Nairobi. Maxwell, D. and Caldwell, R. (2008) The Coping Strategies Index: Field Methods Manual, CARE. National Statistical Office (2012) Integrated Household Survey : Household Socio Economic Characteristics Report. Zomba, Malawi: National Statistical Office. Ricker-Gilbert, J. (2011) Household-level Impacts of Fertilizer Subsidies in Malawi. PhD Thesis, Michigan State University. Ricker-Gilbert, J. and Jayne, T. S. (2010) What are the Enduring Effects of Fertilizer Subsidy Programs on Recipient Farm Households? Evidence from Malawi. Paper presented at the African Association of Agricultural Economists Meeting. September 19 23, 2010; Cape Town, South Africa (power point). Ricker-Gilbert, Jacob., Jayne, Thom. J. and Chirwa, Ephraim. W. (2010) Subsidies and Crowding Out: A Double-Hurdle Model of Fertilizer Demand in Malawi, American Journal of Agricultural Economics, 93 (1), Centre for Agric. Res. & Dev, Malawi 25

32 Annex 1.2: BACKGROUND PAPER 2: WHAT ARE THE FARM-LEVEL IMPACTS OF MALAWI S FARM INPUT SUBSIDY PROGRAM? A CRITICAL REVIEW Rodney Lunduka a+, Jacob Ricker-Gilbert b, and Monica Fisher c a International Institute for Environment and Development, Gray s Inn Road, London, UK, WC1X 8NH. b Department of Agricultural Economics, Purdue University, 403 W. State Street, West Lafayette, USA, IN c International Maize and Wheat Improvement Center Ethiopia Office, c/o ILRI Sholla Campus, P.O. Box 5689, Addis Ababa, Ethiopia. ABSTRACT This article provides a critical analysis of the current frontier of research evaluating Malawi s Farm Input Subsidy Program (FISP), whose main objectives are increasing maize production, promoting household food security, and enhancing rural incomes. We focus on farm-level studies in Malawi, identifying consistent and contrasting research results in order to draw important policy lessons and provide suggested avenues for future research. While national production estimates suggested dramatic increases in maize production and productivity during the years of the FISP, the farm-level studies found relatively modest increases in maize production and yields over the same period. Consistent with the farm-level results of modest maize production increases, there has been a relative increase in real maize prices and the country continued to import maize during most of the subsidy program years. Furthermore, there is evidence that better-off households gained substantially more than poorer households when they participated in the program. Together these findings cast some doubt on the FISP s ability to reduce food insecurity and poverty. We propose a number of policy lessons and suggestions for rigorous investigation, including research that directly measures the causal impacts of the FISP program on poverty in Malawi 1. Introduction Agricultural input subsidies have been a longstanding contentious strategy of both governments in sub-saharan Africa (SSA) and their development partners to promote agriculture and food security. Many countries in SSA, including Malawi, Kenya, Mali, Zambia, Nigeria, Tanzania, Senegal, Ghana, and Ethiopia, have recently implemented input subsidies for chemical fertilizers and modern seed varieties. In many cases, these programs were replacements of agricultural input subsidy programs that were phased out in the late 1980s and early 1990s as part of structural adjustment programs. 5 In theory, by reducing costs, input subsidies should increase input profitability and reduce farmers financial capital constraints, thus encouraging adoption of modern inputs to boost production. However, the costs of implementing large-scale input subsidy programs may outweigh the benefits in the long run, as funding is directed away from other agricultural investments that may have more potential to contribute to sustainable agricultural development (Jayne et al., 2003; Fan et al., 2008). The present study provides a critical analysis of the current frontier of research evaluating Malawi s 5 In many African countries, the new subsidy programs were quite different from the earlier programs. The former programs, for example, tended to be universal price subsidies with little or no targeting. Farm Input Subsidy Program (FISP). We consider the program s performance in achieving the government s official objectives, which are 1) to increase maize production; 2) promote household food security; and 3) enhance rural incomes, recognizing that these objectives are unlikely to be achievable with the FISP alone. The paper briefly examines the economic costs and benefits of the FISP, given the implications for the program s sustainability. In addition, the study identifies key gaps in the current literature and provides suggested areas for future research. Malawi s FISP is a crucial case study among the African countries that have reintroduced subsidies, because Malawi has received wide recognition and has been hailed in the press as the site of the first African green revolution (Dugger, 2007; Denning et al., 2009; Sachs 2012). As such, Malawi s input subsidy program serves as a model that other countries look at when designing their own subsidy policy. Investigating the impacts of Malawi s input subsidy program leads to interesting and sometimes conflicting findings. For example, official government reports from Malawi indicate that the program has increased the use of modern maize seed (hybrid and open pollinated varieties) and chemical fertilizer by smallholder farmers. Along with favourable rains, this has resulted in increased maize production and productivity, and subsequent improvements in national and household food security (Dorward and Chirwa, 2011). However, maize prices increased during the years of the subsidy program, which is counter-intuitive to what we would expect during years of good harvests, Centre for Agric. Res. & Dev, Malawi 26

33 although the maize price increases in Malawi could have been related to the 2008 world price spikes (Figure 1). In addition, the program has not escaped criticism on its effectiveness and efficiency in raising maize productivity, its impacts on the development of sustainable commercial input markets, its opportunity costs in terms of crowding out other investments, its overall return on investment, and its sustainability (Ricker-Gilbert et al., 2013; Holden and Lunduka, 2012b). To date, research has measured the impacts of Malawi s FISP on 1) maize productivity; 2) the development of sustainable commercial input markets; 3) crop diversification; 4) application of other inputs; 5) natural resource use; 6) the malefemale farmer disparity in adoption of modern maize; and 7) household economic well-being (Table 1). Studies have also examined program design and implementation issues (Dorward and Chirwa, 2011; Holden and Lunduka, 2012a), and measured the program s opportunity costs in terms of crowding out of other investments and its overall return on investment (Ricker-Gilbert, et al., 2011; Holden and Lunduka, 2012b). These impact assessments have used different data sets and different methodologies, and at times have reached different conclusions (Ricker-Gilbert et al., 2013). The present paper builds on previous research that evaluated input subsidy programs in Malawi. Harrigan (2008) discussed the strengths and weaknesses for reducing poverty and food insecurity of two earlier subsidy programs in Malawi the Starter Pack Scheme (SPS) and the Targeted Inputs Program (TIP). SPS and TIP evaluations suggested that both programs increased maize production, and that maize surpluses subsequently led to a reduction in the price of maize, further improving food security for the large proportion of Malawi s farm households that are net buyers of food (Levy, 2005; Levy and Barahona, 2002). Dorward and Chirwa (2011) provided conceptual and logistical information on the Malawi FISP between 2005/06 and 2009/10. The present article examines the empirical evidence from farm-level studies on the Malawi FISP, identifying consistent and contrasting research results in order to draw important policy lessons and guide future research. We focus on farm-level studies because the bulk of the impact assessments have been done at this level. Research synthesized in this article has relevance beyond Malawi, and should be particularly applicable to some of the other African countries (e.g., those with similarities to Malawi in terms of land availability and poverty and where the staple crop is fertilizer responsive and tradable) that have implemented or are considering starting similar agricultural input subsidy programs. The remainder of the article is organized as follows. The next section provides essential background on the FISP. We then discuss some key design and implementation issues that challenge the program s effectiveness and efficiency. The fourth section summarizes evidence on the farm-level impacts of the FISP. The paper concludes with a discussion of the policy lessons learned from the FISP experience, the strengths and limitations of existing work on the program s micro-level impacts, and the most pressing research questions to be addressed in future studies. 2. The Malawi Farm Input Subsidy Program 2.1 The program background from Smaller input subsidy programs have existed for decades in Malawi, however the Government of Malawi implemented the large-scale FISP in the 2005/06 cropping season with the official aims of increasing maize production, promoting household food security, and enhancing rural incomes. The program targets approximately 50% of farmers in the country (80% of smallholder farmers) to receive subsidized fertilizer for maize production, with additional vouchers for tobacco fertilizers and for modern maize seeds. Malawi largely funds FISP through its national budget. In 2005/06, the program accounted for 5.6% of the national budget and this increased to 16.2% of the national budget in 2008/09. In 2010/11, the program accounted for 6.5% of the national budget (Table 2). It is important to note that the national budget in Malawi receives considerable donor support. For example, in 2007/08 budgetary support from donors represented 43.7% of Malawi s total proposed national budget (Norwegian Agency for Development Cooperation, 2007; Government of Malawi, 2007). Table 2 presents important features of the program for 2005/06 to 2010/11. Under the program, coupons for subsidized fertilizer were distributed each year to about 2 million of Malawi s 2.5 million smallholder farm households (MoAFS, 2008). In its second year, 2006/07, the program began distributing coupons that could be redeemed for free bags of modern maize seed. The modern maize seed was free, but the quantity of seed provided by the program was small and differed depending on seed type: 2 kg for hybrid seed and 4-5 kg for open pollinated varieties (OPVs). The aim was to provide smallholder maize farmers with an opportunity to use and learn about modern seed. In 2008/09, the subsidy program was extended to cash crops, namely tobacco and cotton. In 2009/10, FISP concentrated on maize fertilizers, maize seed, legume seed (groundnuts, pigeon peas, soya beans, common beans), and storage pesticides. No cash crops, except cotton, were included in the program that year. In 2005/06 the program procured 131,388 metric tons (MT) of subsidized fertilizer sales at a cost to Government of US$55.7 million, equivalent to 2.2% of GDP or 2% after deducting donor contributions (Minde et al., 2008). The redemption price for fertilizer was US$7.35 per 50kg fertilizer bag, which represented a 64% subsidy to the Centre for Agric. Res. & Dev, Malawi 27

34 beneficiary farmers. In the following two years, the total amount of subsidized fertilizer was increased to 179,000 MT in 2006/07 and to 216,553 MT in 2007/08. In addition to the fertilizer subsidy, the Government also subsidized 4,524 MT and 5,541 MT of modern maize seed in 2006/07 and 2007/08, respectively. The total cost of the program increased from 5.6% (2005/06) to 8.9% (2007/08) of the total national budget. In 2008/09, the total amount of subsidized fertilizer was reduced from the previous year to 202,278 MT. However due to the increase in fertilizer prices on the world market, the total cost of the program almost doubled to US$274.9 million, about 16.2% of the total national budget. The total amount of subsidized fertilizer in 2009/10 was 160,000 MT. As the world fertilizer prices were reducing after the increase in 2008 and the improved control of subsidy volumes by the Government (Dorward and Chirwa, 2011), the total program cost reduced substantially from the previous year to US$114.6 million or 8.2% of the total national budget. The subsidized fertilizer redemption price for 2009/10 was reduced from the previous year to US$3.34. Since the commercial price for a 50 kg bag of fertilizer was about US$67 in 2009/10, this implied a subsidy of US$63.66 per bag or 95% of the commercial price. In 2010/11, 160,000 MT of fertilizer was distributed as in 2009/10, but the total program cost increased slightly to US$127.5 million, due largely to rising fertilizer prices and increased volumes of subsidised seed. As shown in Table 2, program costs have generally risen over time, from US$55.7 million in 2005/06 to US$127.5 million in 2010/11, with the peak in program costs of US$274.9 million occurring in 2008/09. Most of the program cost increase was in the early years and, since 2008/09, program costs have declined or been stable. As a percentage of the national budget, the program increased slightly from 5.6% to 6.5% between 2005/06 and 2010/11, with higher percentages of the national budget being spent on the program in the intermediate years, particularly in 2008/09 which was an exceptional year with high world fertilizer prices. Given the high cost of the program, observers have questioned the effectiveness, efficiency, and sustainability of FISP spending compared to spending on other possible agricultural and developmental investments. 2.2 Costs and Benefits of FISP Given the high cost of implementing the program and the increase in maize prices that has accompanied the FISP (Figure 1), one is compelled to question the economic returns of the program at both farm and national level. A first question is whether or not inorganic fertilizer use for maize is profitable and under what conditions. This is extremely important to understand the viability of the program. Sauer and Tchale (2009) found that maize was marginally profitable when using only organic fertilizer given market input and output prices, but the profitability increased when the inorganic fertilizers were combined with other soil fertility-enhancing inputs like organic manure. Recent evidence suggests that the returns to subsidized fertilizer are not symmetric across the distribution of smallholders. Ricker- Gilbert and Jayne (2012) looked at the distributional returns to maize production in Malawi using quintile regression. They found that at the 10 th percentile of the maize production distribution, one kg of subsidized fertilizer acquired by a household contributes just 0.86 kg of maize. This compares to a response rate of 5 kg of maize per kg of subsidized fertilizer at the 90 th percentile of the maize production distribution. This finding indicates that limited resource farmers get a lower return from inorganic fertilizer, due to limited resources, limited management ability, and poor soil quality, than do better-off farmers. Understanding the full economic benefit of the FISP requires consideration of the direct effects of the program on subsidized coupon recipients and the subsequent multiplier effects on household livelihoods and the rural economy (Dorward et al., 2008). However, we have not found any study that has directly assessed the multiplier effects of the subsidy program in Malawi. Dorward and Chirwa (2011) estimated the direct benefits of the program by calculating the benefit-cost ratios from 2005/06 to 2008/09. Their calculations were based on assumed maize-nitrogen response rates of 18 (high), 15 (medium), and 12 (low) kg of grain per kg of nitrogen and the prevailing commercial fertilizer prices. 6 They showed that benefit-cost ratios ranged from 0.09 to 1.54, with a mid-estimate of In another study using different price and yield scenarios for 2011/12, Dorward and Chirwa (2012) estimated that benefit-cost ratios range from 1.1 to 1.7 without any allowance for multipliers from growth linkages, and from 1.2 to 1.9 with allowance for growth multipliers, with the most likely estimates roughly in the middle of these ranges. Ricker-Gilbert and Jayne (2012) estimated the plot-level conditional mean response of maize on fertilizer using a firstdifference method on both total and subsidized fertilizer used by households. They found that, on average, the response of maize is 2.71 kg grain per kg of subsidized fertilizer acquired by households. 7 Using the maize-fertilizer response rates from Ricker-Gilbert and Jayne (2012) and government program costs per kg of fertilizer 6 Grain to Nitrogen response ratios can be divided by 3 to compare with grain to fertilizer ratios. 7 The subsidized fertilizer to maize response rate in Ricker-Gilbert and Jayne (2012) accounts for an estimated 18.0% average crowding out rate of commercial fertilizer by subsidized fertilizer. Centre for Agric. Res. & Dev, Malawi 28

35 distributed through the FISP from Dorward and Chirwa (2011), we further calculated the benefit-cost ratio of using fertilizer at prevailing maize prices, FISP costs, and estimated response rates (Table 3). Some caveats related to the calculations in Table 3 deserve mention. First, the three benefit-cost ratio estimates in the table exclude the administrative cost of the program that would be part of the costs in column B. These costs include administrative and personnel costs for the Ministry of Agriculture and Food Security (MoAFS). Second, the calculations in Table 3 assumed the same output and input prices across Malawi locations. However, maize output prices did not increase at the same rate as fertilizer costs between 200/06 and 2008/09 (Table 3 column A and B). The results show that the benefit-cost ratios depend greatly on the cost of fertilizer to the government. We see from column D that when we use the assumed maize-to-fertilizer response ratio of 5:1, the benefit-cost ratios are greater than 1 in all years of the program. However, when we use the estimated marginal response to a kg of subsidized fertilizer from Ricker-Gilbert and Jayne (2012), we find that the benefit-cost ratio is below 1 in all years except for 2007/08, when it is slightly above 1 at 1.15 (column E of Table 3). When we value subsidized fertilizer at the price recipient farmers pay for the subsidized fertilizer, the benefits are well above 1 (column F of Table 3). This indicates that households who received subsidized fertilizer likely benefited from it, which is not surprising given that these households acquired a productive input at up to a 90% subsidy rate. However, the findings in column E of Table 3 cast doubt on the program s sustainability, given that the returns to subsidized fertilizer do not generate enough revenue to cover the cost. 8 A simple explanation of the scenario generated in Table 3 is that if benefit-cost ratios presented in columns D and E are greater than 1, then returns to a kg of subsidized fertilizer are larger than giving a household cash; a benefit-cost ratio equal to 1 indicates that a kg of subsidized fertilizer is equal to giving recipients cash; a benefit-cost ratio less than 1 indicates that the returns to subsidized fertilizer are lower than giving recipients cash. From column E it appears that the subsidy program would have provided nearly equivalent or lower returns than providing recipients with cash, while the response rate in column D makes the program look more favourable. 3.0 FISP Design and Implementation Issues 8 The results in Table 3 were similar when we valued the cost of the fertilizer (column B of Table 3) at its average commercial price in the market during each year. The annual average commercial prices in US$/kg of fertilizer were ($0.49 in 2005/06; $0.49 in 2006/07; $0.59 in 2007/08; and $1.25 in 2008/09). Under the Malawi FISP, beneficiary households receive subsidized input coupons from the government that can be redeemed at designated outlets throughout the country. Since FISP s establishment in 2005/06, the coupons have been officially targeted toward the productive poor, although the definition of who these households are has changed somewhat over the program years. This is discussed in more detail in section 3.3 below. The Ministry of Agriculture and Food Security (MoAFS) issues and distributes the coupons through their local staff at the extension planning areas (EPAs) 9, in collaboration with village chiefs and Village Development Committees (VDCs) who identify beneficiary households at the village level. A beneficiary is entitled to two fertilizer coupons for use in maize production: one 50 kg bag of 23:21:0+4S fertilizer and one 50 kg bag of Urea fertilizer. Beneficiaries are also entitled to receive a maize seed coupon that can be redeemed for 2 to 5 kg of modern maize seed, either hybrid or OPV, for free. While the FISP has generally been successful at getting subsidized fertilizer and seed to rural areas, implementation of the program has faced four main challenges: politicization, diversion and leakages, targeting problems, and timely delivery of fertilizer to the local outlets Politicization of input subsidies Few would dispute that input subsidies in Malawi are highly political. After winning the election in May 2004, the late president Bingu wa Mutharika, made a bold step by reintroducing largescale agricultural subsidies in the 2005/06 growing season, after a period with a small-scale input subsidy program (the SPS, and later the TIP that were introduced in the late 1990s and early 2000s). Mutharika was directly involved with the FISP design and implementation throughout his time in office. It has been argued that Mutharika used the FISP for political motives to stay in power, which undermined technical improvements to the subsidy program (Chinsinga, 2013). Empirical evidence from Malawi supports and quantifies the notion that the FISP program is politicized. Mason and Ricker-Gilbert (2013) found that, on average, households in districts where the ruling party won the last presidential election received 1.66 kg more subsidized seed and kg more subsidized fertilizer than households in other districts. Holden and Lunduka (2012a) similarly found that households in districts where the ruling 9 The Extension Planning Area (EPA) is the smallest operational zone, which is manned by an extension officer in the Ministry of Agriculture and Food Security. The EPAs fall within a district and each district has two or more EPAs. The demarcation generally follows the ecological landscape of the district. Centre for Agric. Res. & Dev, Malawi 29

36 party had a large number of supporters had a higher probability of receiving fertilizer coupons compared to districts where the ruling party had fewer supporters. These findings are important because politicization of input subsidies has far-reaching ramifications that influence the ability of subsidy programs to meet the stated objectives of boosting staple crop production and reducing poverty Diversion and Leakage In addition to politicization, the Malawi FISP has been criticized for widespread diversion and leakage (Dorward et al., 2008; Holden and Lunduka, 2010a). Diversion refers to both coupons for subsidized fertilizer and subsidized fertilizer that is taken by government officials and resold as commercial fertilizer. Leakage refers to coupons and subsidized fertilizer that is resold by recipient households on the secondary market. There is evidence to suggest that diverted and leaked coupons have found their way to smallholder farmers who purchase coupons in the illegal market or access cheap fertilizer sourced from FISP. This could positively impact maize production if farmers who acquire diverted or leaked fertilizer use it more efficiently than farmers targeted by the program, and smallholder famers may still access cheap fertilizer through this illegal market. However, evidence shows that better-off individuals who are less cash constrained often acquire diverted and leaked fertilizer. To the extent that these households use the diverted and leaked subsidized fertilizer in place of commercial fertilizer, then diversion and leakages contribute to crowding out of commercial fertilizer, which reduces the total amount of new fertilizer that enters the system and ultimately ends up on farmers fields. We discuss in detail the effect of this crowding out of commercial fertilizer in section 4.4. The evidence shows that there has been a substantial illegal market for coupons and subsidized fertilizers, especially in the first few years of the program (Dorward and Chirwa, 2011). Diversion and leakages have led to the birth of secondary markets for coupons and fertilizers, and this has resulted in fewer subsidized input coupons reaching the villages than was intended. Using their survey data for 2007 and 2009, Holden and Lunduka (2012a) found that as much as one-third of the subsidized fertilizers may have been diverted or leaked into this secondary market. On average, a household in their sample of 450 farm households purchased 0.23 bags of subsidized chemical fertilizer. Sample households reported that about 20% of their total subsidized fertilizer came from illegal sources. These households either bought fertilizer coupons from someone else, or redeemed the coupons at official distribution outlets, or they bought cheap commercial fertilizer that was likely diverted subsidized fertilizer directly from vendors. For the purposes of this study, we quantified the diversion estimates in Malawi for the 2009/10 agricultural season using the nationally representative Integrated Household Survey III (IHS3). We followed the method presented in Mason and Jayne (2013) that the authors use to calculate diversion/leakage for Zambia. The method involves simply summing the total quantity of subsidized fertilizer that households in the survey said that they acquired in 2009/10, and dividing that number by the actual amount of subsidized fertilizer that the government said it distributed in that year, which was 160,000 MT. 10 The diversion rate is one minus the ratio of actual receipt/official receipt. Using this formula, our diversion estimate was 42% (93,172 tons of subsidized fertilizer actually received by farmers) after we adjusted outlier observations for household who reported acquiring greater than 600 kilograms of subsidized fertilizer. This quantitative diversion estimate for Malawi in 2009/10 is consistent with Holden and Lunduka s observation of 33% diversion, and Mason and Jayne s average diversion rate estimates between 30% and 40% in Zambia. Many have questioned how these illegal fertilizer coupons reached the markets. There is some evidence indicating the illegal coupons and subsidized fertilizer are not mainly from leakages where households that received coupons then sold them. Research suggests that households that received coupons preferred to keep and use the coupons themselves rather than sell them. For example, Holden and Lunduka (2013) conducted a real choice experiment between cash and fertilizer and found that more than 80% of the households who were offered a similar package of the subsidy for free, preferred to keep the package even when they were offered a willingness to accept (WTA) price of MK 9,000 (equivalent to the commercial price). In another survey, 7%-8% of the sampled households were presented with offers to sell their coupons, but only 1% of them reported having sold coupons (Holden and Lunduka, 2012b). Similarly, Ricker- Gilbert et al. (2011) found that of the 716 sampled respondents who reported receipt of subsidized fertilizer coupons in 2006/07, only 12 (1.6%) said that they sold the coupons, and one respondent reported having given the coupon away. While it is possible that respondents in these two surveys were unwilling to disclose coupon sales, given the activity is illegal, if survey results are in fact accurate, then the large majority of the coupons in the unofficial market were likely to have come from sources other than FISP beneficiary households. It is plausible that officials with control of subsidized inputs, such as village chiefs, traditional authorities 10 The IHS3 dataset allows us to estimate nationally representative totals of subsidized fertilizer using population weights that have been constructed by Malawi s Central Statistical Office. Centre for Agric. Res. & Dev, Malawi 30

37 and government officers, have been largely responsible for the widespread diversion of coupons and subsidized fertilizer. The first possible source of diversion is the manner in which the national list of potential beneficiary households has been calculated. This list is likely unreliable and it may have been inflated from 2.5 million (according to the 2008 population census) to 3.8 million households, so that many non-existing households received fertilizer coupons (Dorward et al., 2010; Holden and Lunduka, 2012b). These unallocated coupons were available to some government officials and traditional leaders who could have introduced them on the market. Another source of diversion is that, although records are kept for the total amount of subsidized fertilizer distributed by type, no proper record exists for the number of fertilizer coupons printed and distributed for the supplementary subsidy distribution (Logistic Unit, 2009). This makes the administrative coupon distribution system vulnerable to illegal activities and diversion, because it is not possible to balance the total number of coupons printed and distributed with the total amount of subsidized fertilizer distributed (i.e. the first distribution and the supplementary vouchers) Targeting issues Another challenge that the Malawi FISP has faced is the targeting of intended beneficiaries. As stated earlier, the program is intended to target the productive poor. In 2005/06 and 2006/07, coupons were allocated to districts on the basis of proportional maize area. Within villages, input subsidy program committees were supposed to identify the productive poor beneficiaries, defined as full time smallholder farmers who cannot afford to purchase one or two bags of fertilizer at prevailing commercial prices, as determined by the leaders in their areas (Dorward et al., 2008, p. 23). Since 2007/08, the productive poor have been defined as farm households with the land and human resources to use the subsidized inputs, but without the financial capital to purchase inputs at commercial prices (MoAFS, 2008). These definitions of the productive poor can be compared to the official targeting criteria for beneficiary selection under FISP as of 2007/08: 1) households headed by a Malawian who owns and currently cultivates land; 2) vulnerable households, including guardians of physically challenged persons, and households headed by females, orphans, or children; and 3) only one beneficiary per household, the household head (MoAFS, 2008). There is some inconsistency between the productive poor targeting and the official targeting criteria, since vulnerable households often do not have the land and human resources to productively use subsidized inputs, which complicates the evaluation of targeting effectiveness. Research has not investigated whether or not the productive poor have been the primary recipients of FISP coupons, but empirical evidence indicates that the targeting of coupons to vulnerable households, such as poor or female-headed households (FHHs), has not been successful in many years of the subsidy program. For 2005/06 and 2006/07, Holden and Lunduka (2012a) found for their sample of 450 households in six districts of Malawi that poor households were frequently excluded, while betteroff households were more likely to have received coupons. Chibwana et al. (2012a) found for sampled households (n = 380) in Kasungu and Machinga districts in 2008/09 that female-headed households were less likely than male-headed households to receive coupons and that asset-poor households were also less likely than better-off households to receive coupons. Ricker-Gilbert et al. (2011), using a nationally-representative sample for 2003/04 and 2006/07 (n = 2,406 households), also found that coupons were disproportionately targeted to maleheaded households and those owning relatively more physical assets, including land. Fisher and Kandiwa (2013) using the nationally representative Malawi Integrated Household Survey 3 (IHS3) for 2010/11, similarly found that household wealth level and agricultural landholding positively correlated to FISP coupon receipt. However, contrary to other research cited above, the study found that female-headed households were more likely than male-headed households to receive FISP coupons. This may indicate improved targeting of female-headed households in recent years, but research indicates a need to ensure that poor households are key beneficiaries of the program. The observed administrative targeting problems described above are thought to be the result of several factors. First, due to the diversion of coupons and subsidized fertilizer out of the system, too few coupons reached the villages, with a typical response of village leaders being to reduce the number of coupons per beneficiary household from two to one or none (Holden and Lunduka, 2010a). Second, within villages, targeting criteria have often been ignored due to a culture of egalitarianism that prevails in rural Malawi. As a result, fertilizer and maize seed coupons were often divided among households of various socioeconomic status, rather than given preferentially to the poor (Holden and Lunduka, 2012a). A somewhat even distribution of the coupons to households of various socioeconomic situations could result in a higher rate of coupon use by the non-poor versus the poor, because the coupons require some cash to redeem the subsidized fertilizer, which may prevent the very poor from redeeming coupons. Elite capture of coupons is a third plausible explanation for why targeting of FISP coupons to vulnerable households has not worked well (Holden and Lunduka, 2010a). A study of an agricultural input subsidy program in Tanzania found that elected village officials were more likely to receive subsidized input coupons and these officials also Centre for Agric. Res. & Dev, Malawi 31

38 tended to be less poor than the general population (Pan and Christiaensen, 2012) Issues with the timely delivery of FISP coupons A fourth logistical problem faced by the Malawi FISP is that in the initial years of the program, coupon distribution was completed later than is agronomically optimal, although there is evidence of improvement in this regard in recent years. Kelly et al. (2010) note that most stakeholders in 2008/09 were of the view that the announcement of tenders improved, but the tenders were awarded late. Dorward and Chirwa (2011) reported that in 2008/09 finalization of the seed supply contracts was done by the end of November, while about 68% of the fertilizer sales were completed by the end of December. That left about 30% of fertilizer sales to be done in January. Since maize planting begins as early as October, by the time many beneficiary farmers received their coupons, it was too late to plant the maize seed and/or their maize crop had already matured to the point that the application of fertilizer was too late for potential yield gains to be achieved. The failure to deliver the input coupons on time has negatively impacted the program s efficiency because maize yields are sensitive to the timing of fertilizer application (Minde et al., 2008; Xu et al., 2009). On a positive note, in recent years fertilizer distribution contracts have been awarded early enough to allow fertilizer deliveries to depots and markets just before the growing season (Dorward and Chirwa, 2011; Holden and Lunduka, 2010a). 4. Farm-level Impacts of FISP This section reviews the evidence on farmlevel impacts of FISP. We discuss the empirical findings of 13 studies of FISP s impacts on maize production and yield, maize prices, cropland allocation, demand for commercial fertilizer and organic manure, the differential rates of modern maize adoption by female and male farmers, and farm household economic well-being (Table 1) Maize production and maize yield Since FISP s inception in the 2005/06 cropping season, official government statistics show that Malawi has had record-level maize production (Figure 1). In the program s first year, total maize production was estimated at more than double the 2004/05 harvest, with the nation producing an estimated surplus of 510,000 MT above the national maize requirement. The 2006/07 harvest was officially estimated at 3.44 million MT, generating a surplus of about 1.34 million MT (Denning et al., 2009). In the 2011/12 cropping season, the Malawi Vulnerability Assessment Committee (MVAC) estimated that production was 3.9 million MT, translating to a surplus of 1.2 million MT (GOM, 2012). The estimates cited above suggest dramatic increases in maize production under the FISP. However, there is reason to exercise caution in accepting these figures, because Malawi observers have increasingly questioned the reliability of the national maize production estimates. For example, the 2007/08 official maize production figure indicated a maize surplus, but it is widely believed that the 2007/08 maize harvest was over-estimated by at least 25% (Jayne et al., 2008). Furthermore, data on informal trade flows indicate that Malawi remained a net importer of maize during many years of the FISP. FEWSNet (2013) reported that the country net-imported 52,500 MT of maize in the 2007/08 marketing season through informal crossborder trade flows. Maize net imports of nearly 62,000 MT were also recorded in the 2008/09 season, and 48,500 MT in 2009/10. Malawi s informal terms of trade switched between 2010/11, 2011/12, and 2012/13, as the nation became a net informal exporter of maize according to FEWSNET (see Table 4). However, due to the nature of the informal trade, FEWSNET may not have been able to capture all the trade across the borders hence their figures are just indicative and not final. Another issue to take into account is that local conditions around the border areas may dictate the patterns of trade regardless of production deep inside the country. Hence, informal cross border figures must be carefully interpreted. Studies using farm household-level data suggest that the FISP has had a positive and statistically significant impact on maize production, but the magnitudes of the maize production increases from the household data are considerably smaller than official government statistics indicate. 11 For example, Ricker-Gilbert and Jayne (2011) used six years of data on fertilizer use to estimate how acquiring subsidized fertilizer over time impacts maize production and other indicators. Results showed that an additional kg of subsidized fertilizer received in the current year boosted maize production by 1.82 kg in that year, on average. As well, an additional kg of fertilizer acquired by households in each of the three previous years was found to boost maize production by 3.16 kg in the current year, on average. The increase in maize production from past receipt of subsidized fertilizer could be due to nutrient build up in the soil, or a learning and experimentation process. Further disaggregating maize into modern and local varieties, maize production effects were found to be higher for modern maize varieties. 11 It should be noted that studies using farm-level survey data are also subject to measurement error that may bias their results. The level of measurement error is likely to differ across datasets depending on a number of factors including how the data on agricultural production and plot/farm sizes were collected. Centre for Agric. Res. & Dev, Malawi 32

39 Chibwana et al. (2010) found the average gain in maize yield attributable to receipt of a complete package of coupons, i.e. both seed and fertilizer coupons, was 447 kg/ha, on average. The gain from accessing fertilizer only, i.e. no modern maize seed, was 249 kg/ha. Holden and Lunduka (2010b) also found a significant positive effect on maize yields as a result of access to subsidized fertilizer, but they did not report the magnitude of the estimated effect. Interestingly, the latter study also found that targeted households had significantly lower maize yields than those not targeted by the program, whether receiving subsidies or not. This was found by assessing the yields of households that had been erroneously excluded in the program based on participation predictions. It may be that the vulnerable households targeted by the program are less efficient farmers or are constrained from using fertilizer efficiently due to a lack of resources. Holden and Lunduka (2010b) provided insights on subsidized fertilizer responsiveness of both hybrid and local maize. The study found that plots that did not receive subsidized fertilizer but used commercial fertilizer had the highest fertilizer response rate. Local maize had the lowest fertilizer response rate. Interestingly, plots on which subsidized fertilizer was applied to hybrid maize had lower fertilizer responsiveness than hybrid maize plots on which commercial fertilizer was applied. The latter finding is unexpected since there is no difference between subsidized and commercial fertilizer. A plausible explanation for the observed lower fertilizer response rate for subsidized fertilizer is the late distribution of fertilizer under FISP, which was discussed earlier. Acquiring fertilizer on time was found to roughly double the maize response rate to fertilizer in Zambia (Xu et al., 2009) Maize prices As mentioned, national-level estimates indicate that maize production has increased in Malawi since the implementation of FISP in 2005/06, and micro-level studies confirm positive, although modest, growth in maize production and productivity. Thus, it is interesting and curious that maize prices in Malawi have increased in a number of recent years from just below US$100/ton in 2005 to US$400/ton in 2010 (Figure 1). In a completely closed economy, we would expect increased maize production to exert downward pressure on maize prices. In a relatively open economy, like that of Malawi with cross-border trade, domestic maize prices may be influenced by world prices. Recent evidence on spatial market integration and spatial price transmission from southern and eastern Africa generally indicates that maize markets are currently fairly integrated and are becoming more integrated and efficient over time (Tostau and Brorsen, 2005; Awudu, 2007; Myers, 2008; Burke 2012; Myers and Jayne 2012). However, there may be certain markets in Malawi and in other countries in the region that are not well integrated into the broader region and where local production may affect local maize prices. Figure 1 shows that retail maize prices in Malawi responded to changes in production fairly consistently until 2008/09 when prices spiked and continued to rise for the next few years. Dorward et al. (2010) argued that this trend of rising maize prices could be due to a number of factors including the exportation of maize, the purchase of maize by the Malawi government for its strategic grain reserve (SGR), higher welfare and real incomes following the 2005/06 harvest, changes in informal cross-border flows, over-estimates of national maize production following the implementation of the subsidy program, and storage of maize by traders. Several of these factors could have reduced the supply of maize at markets and thereby increased the price of maize. In 2007/08, roughly 303,000 MT of maize were exported to Zimbabwe. In 2009/10, about 130,000 MT were purchased by the government for the SGR, and a further 100,000 MT were estimated to have been bought and held in storage by private traders (FEWSNet, 2009; Jayne et al., 2010). Dorward et al. (2010) also explained the rising maize price phenomenon as being caused by large storage losses as a result of increased production of hybrid maize, which is more susceptible to storage pests than local maize. Although government speculated that storage losses in 2007/08 were greater than 30%, Jayne et al. (2010) estimated that such losses were about 14%. The latter figure agrees reasonably well with the 12.9% figure by a Government of Malawi survey of grain storage losses which was conducted in 2005 prior to FISP implementation. This suggests that the storage losses explanation for rising maize prices is unlikely to be valid. Recent empirical evidence estimating market-level impacts of fertilizer subsidies on maize prices suggests that the program has had a marginally statistically significant and small economic effect on maize prices in Malawi. Ricker-Gilbert et al. (this issue) found that roughly doubling the size of Malawi s subsidy program, by increasing the average amount of subsidized fertilizer distributed to each district by 4,000 MT per year, would only reduce maize prices by 1.2% to 1.6%, on average. As the majority of small farmers are net consumers of maize, and poor urban consumers in Malawi spend a significant share of their income on maize, the effect of FISP on maize prices has important welfare implications for the vast majority of Malawi s population. However, based on the findings from the aforementioned study, input subsidy programs likely cannot be justified based on large downward effects on maize prices Land allocation decisions Previous input subsidy programs in Malawi, the SPS and TIP, were criticized for creating and perpetuating widespread dependency on maize (Harrigan, 2008). Although cereal specialization can Centre for Agric. Res. & Dev, Malawi 33

40 benefit a farm household by increasing its calorie availability and income, it can also introduce tradeoffs in terms of maintaining soil fertility and managing weather-related agricultural risks. Simplified cereal-based cropping systems with minimal inputs are associated with severe reductions in soil nutrients (Snapp et al., 2002). Moreover, monoculture cropping systems may increase the vulnerability of farmers to climate variability and change. Farmers in Africa have historically diversified their cropping systems to self-insure against effects of adverse weather, as different crops are affected differently by weather events (Adger et al., 2003; Di Falco et al., 2010). Studies by Chibwana et al. (2012) and Holden and Lunduka (2010b) provide somewhat different evidence on whether Malawi s FISP has influenced farmers decisions to simplify or diversify their cropping patterns. Chibwana et al. (2012) found that receipt of agricultural input subsidies under the FISP was associated with crop simplification: on average, sampled farmers who received coupons for seed and fertilizer allocated 16% more land to maize than those who did not. Furthermore, the increased share of a household s farmland allocated to maize occurred at the expense of other crops (legumes, cassava, and sweet potato), which were allocated 21% less land on average. By contrast, Holden and Lunduka (2010b) using a panel dataset for 2006, 2007, and 2009, reported that the total maize area among their sampled households had decreased from 0.73 in 2006 to 0.64 in Holden and Lunduka (2010b) did not directly show that the input subsidy program caused maize area to decrease, but the analysis provided descriptive evidence that when FISP was scaled up, maize intensification could have facilitated crop diversification by releasing maize areas. The different findings on crop simplification vs. diversification of Chibwana et al. (2012a) and Holden and Lunduka (2010b) can be explained as follows. First, the two study contexts are different. Chibwana et al. (2012a) focused on the districts of Kasungu and Machinga, while the study of Holden and Lunduka (2010b) concerned Kasungu, Lilongwe, Machinga, Chiradzulu, Thyolo, and Zomba districts. Second, Chibwana et al. (2012) used the share of maize area as the dependent variable, while Holden and Lunduka (2010b) focused on total maize area. The results of the two studies could in theory be consistent, since the share of maize area could decline while total area (including total maize area) could increase due to area expansion, although it should be noted that area expansion is not feasible in most of Malawi s maize growing areas. Third, Chibwana et al. (2012a) relied on cross-sectional data and the empirical analysis controlled for potential endogeneity of FISP coupon receipt with instrumental variables estimation. Therefore, their study can claim that the subsidy program had a causal effect on the share of maize area, assuming instrument validity. Conversely, while Holden and Lunduka (2010b) used panel data, their descriptive analysis was only suggestive of change in the maize area over time and their finding of maize area reduction cannot be directly linked to the subsidy program. Both studies have the same drawback of relying on small samples that may not be easily extrapolated to the national level. Future research that uses nationally-representative panel data and controls for potential endogeneity of FISP coupon receipt is needed to better assess the relationship between agricultural input subsidies and cropping diversity. Another study of FISP s impact on land allocation examined how the subsidy program impacted forest use in the Chimaliro and Liwonde Forest reserves in Kasungu and Machinga districts (Chibwana et al., 2012b). Results indicated that receipt of subsidized input coupons reduced forest expansion modestly, consistent with Fisher and Shively (2007) who found that the Malawi SPS raised agricultural output without encouraging agricultural expansion. On the other hand, Chibwana et al. (2012b) showed that tobacco subsidies delivered as part of the FISP generated a derived demand for timber (to construct tobacco drying sheds) and therefore may have had a detrimental impact on forest cover. The net effect on forests of the FISP could not be measured with the available data Demand for commercial fertilizer, seed, and organic manure To achieve the objectives of the FISP, there is a need to promote sustainable development of commercial input distribution systems and increase fertilizer use by smallholder farmers. Holden and Lunduka (2013) found that rural Malawian households value fertilizers highly, whether they buy them at commercial or subsidized prices. An important concern is whether or not the subsidized fertilizer distributed under the FISP crowds out commercial fertilizer purchases and the use of organic manure. Using panel data from 2003/04 and 2006/07, Ricker-Gilbert et al. (2011) reported that the current subsidized fertilizer program has had a significant negative impact on farmers commercial fertilizer purchases: on average, each additional kg of subsidized fertilizer reduced farmer purchases of commercial fertilizer by 0.22 kg. This means that each additional kg of subsidized fertilizer contributed an additional 0.78 kg to total fertilizer use after accounting for crowding out. The study also presented evidence that wealthier farmers were more likely than poorer households to crowd out their commercial fertilizer purchases when they acquired subsidized fertilizer. The top fifth of the population displaced 30% of their commercial fertilizer purchases when they acquired subsidized fertilizer, while the poorest fifth of the sample displaced 18% of their commercial purchases when they acquired Centre for Agric. Res. & Dev, Malawi 34

41 the subsidized fertilizer. This finding is as expected, since wealthier households have the resources to purchase fertilizer commercially, and it may suggest that targeting agricultural input subsidies to farmers with limited resources minimizes crowding-out effects. Ricker-Gilbert and Jayne (2012) updated the crowding out estimates from Ricker-Gilbert et al. (2011) using an additional wave of panel data collected in 2008/09. The authors found the average crowding out rate over the three waves declined from 22% to around 18%. This finding indicates that crowding out of commercial fertilizer may have been reduced in the more recent years of the subsidy program. The data suggest that in 2008/09 fewer people acquired extremely large quantities of subsidized fertilizer, greater than 500 kg. This may have been due to the fact that coupon distribution was tightened up in more recent years, as the receipt of vouchers was tied to voter identification, making it harder for people to accumulate more than the official 100 kg of fertilizer per participant. The community-based targeting system that was implemented in recent years may have also helped to steer subsidized fertilizer to households who lacked effective demand for commercial fertilizer compared to earlier years. Mason and Ricker-Gilbert (2013) built on earlier work related to fertilizer crowding out, and estimated the extent to which fertilizer and seed subsidies crowd out commercial seed purchases in Malawi and Zambia. The authors found that in Malawi an additional kg of subsidized maize seed acquired by a household reduced its commercial modern maize seed purchases by 0.58 kg, on average. 12 This finding means that if the government adds 100 kg of subsidized seed to the program, it will only contribute 42 new kg of improved seed to the total use because the other 58 kg will be crowded-out commercial seed. This finding is consistent with the crowding out findings for commercial fertilizer (Ricker-Gilbert et al., 2011), and it indicates that households who received subsidized seed used some of it in place of their commercial purchases. Holden and Lunduka (2012b) investigated whether subsidized fertilizer crowds out organic manure. They found a positive association between fertilizer use intensity and manure use intensity. A 1% increase in fertilizer use intensity was associated with a 1.94% 1.96% increase in the intensity of manure use outside the subsidy program and a 0.62% 1.66% increase in manure use with the subsidy program. A 1% increase in average fertilizer 12 The study by Mason and Ricker-Gilbert (2013) was not able to distinguish between hybrid and open pollinated varieties of maize because the vast majority of sampled farmers referred to all modern maize varieties as hybrids, and did not distinguish between true hybrids and OPVs. price was associated with a 0.43% 0.76% increase in the probability of manure use and a 3.5% 5.3% increase in the intensity of manure use. These results may suggest that some substitutions took place between subsidized fertilizer and manure. However, households that applied commercial fertilizer were likely to apply more manure that households that applies only subsidized fertilizer. However, the small share of Malawian farm plots receiving organic manure can also explain why access to input subsidies, to a small extent, crowded out manure Gender-based impacts of the FISP In SSA and other areas of the developing world, use of modern seed varieties is essential for farmers to increase their crop harvests significantly and improve their living standards (Minten & Barrett, 2008). Nevertheless, certain groups of farmers, notably women, have relatively low rates of adoption of modern crop varieties and other agricultural technologies associated with increased crop yields (Peterman et al., 2010). Targeting coupons to femaleheaded households was not successful in some years of the subsidy program, as highlighted earlier. It is thus important to investigate whether FISP influenced the disparity in agricultural technology adoption between female-headed and male-headed households. Furthermore, since the program distributed the subsidized input coupons to household heads, if male household heads did not share received coupons with their wives, the gap in modern input use between male farmers and female farmers in male-headed households (MHHs) could have increased due to FISP. Two studies explored these issues: Fisher and Kandiwa (2013) for modern maize seed adoption and Chirwa et al. (2011) for the case of fertilizer. Fisher and Kandiwa (2013) provided evidence that FISP did not reduce the modern maize adoption gap between households headed by males vs. females. Using Malawi IHS3 data, they ran a series of simulations based on a logit adoption model to investigate potential FISP impacts on Malawi s gender gap in modern maize adoption. Simulation results showed that targeting FISP coupons to female-headed households (FHHs) was associated with a slight reduction in the gender gap between FHHs vs. MHHs, but the difference was not statistically significant. A plausible explanation for why FISP did not appear to narrow the gender gap for female- and male-headed households, despite the targeting of female-headed households as was reported earlier in this article, is that the FHHs that benefited most from FISP were not poor. The IHS3 data showed that coupons redeemed by asset-poor FHHs averaged US$48 in value, while coupons redeemed by non-poor FHHs had a significantly higher average value of US$59. The non-poor FHHs that received coupons probably would have adopted modern maize in the absence of coupon receipt. The asset-poor FHHs, on the other hand, probably needed Centre for Agric. Res. & Dev, Malawi 35

42 financial assistance, such as a subsidized input coupon, to purchase inputs. In fact, simulations predicted that preferential targeting to asset-poor households could reduce or even eliminate the gender gap in modern maize adoption. As for the modern input adoption gap between male farmers and farmers who are wives in MHHs, both Fisher and Kandiwa (2013) and Chirwa et al. (2011) provided evidence that FISP likely had no effect on the gap. Chirwa et al. (2011) used Agricultural Input Subsidy Survey II (AISS2) data to examine intra-household fertilizer use for plots managed by male and female farmers. 13 The study did not find gender-based bias in intra-household use of subsidized fertilizers: there was no statistically significant difference in subsidized fertilizer application to plots controlled by wives in MHHs and by male householders. The authors explained the result by arguing that most of the subsidized fertilizer was used for production of maize to meet household subsistence needs, and rural Malawian women tend to have stronger bargaining power when it comes to provision of food in the household. Fisher and Kandiwa (2013) similarly found no statistical difference in the monetary value of maize and fertilizer coupons redeemed by wives in MHHs and by male farmers, indicating that male household heads often shared the received coupons with their wives. Interestingly, the use of FISP coupons had no discernible effect on the adoption of modern maize by wives in MHHs, but it did increase modern maize adoption probability among male household heads. Further analysis where current adoption levels were compared with simulated adoption levels for a scenario in which FISP did not exist, suggested the program had no effect on the disparity between the adoption of modern maize by wives in MHHs and by male farmers Household economic well-being The Malawi National Statistical Office (NSO) and the World Bank recently completed a household-level poverty analysis in Malawi, based on nationally-representative Integrated Household Surveys conducted in 2010/11 (IHS3) and in 2004/05 (IHS2) (GOM, 2012b). The report compared IHS3 poverty rates for 2009/10 with IHS2 poverty rates for 2003/04 which was before implementation of the Malawi FISP. The report did not measure the impact of FISP on poverty in Malawi, as the analysis did not control for other changes between 2003/04 and 2009/10 that could have influenced poverty. However, its results provide an important description of household economic well-being, from before and after the program s implementation. 13 The AISS was implemented by Malawi s NSO in 2007 and 2009, and when combined with Malawi s second Integrated Household Survey (IHS2) it is a three-year panel. The data are available upon request. The GOM report found that the overall incidence of poverty in Malawi fell by just two percentage points from 52.4% in 2003/04 to 50.7% in 2009/10. The number of ultra-poor in Malawi rose slightly between those years, from 22.3% in 2003/04 to 24.5% in 2009/10. While urban poverty declined from 25.4% in 2003/04 to 17.3% in 2009/10, rural poverty slightly increased from 55.9% to 56.6% over the same period. The report found substantial inequality among households, and that income inequality, as measured by the Gini coefficient, increased from 0.39 in 2003/04 to 0.45 in 2009/10. The findings from the GOM raise serious questions about whether the FISP has made any substantive contribution towards reducing poverty, and whether the money could have been better spent on other social programs. Addressing these questions is crucial, given that the program cost as much as 16% of the government s budget in some years (Table 2). It should be noted, however, that the GOM s descriptive analysis only looked at the situation before and after FISP, and did not consider the changes in poverty levels with and without FISP. The GOM results, therefore, do not indicate whether or not poverty rates would have been higher had the FISP not been in place. Higher poverty rates in 2009/10 vs. 2003/04 would be consistent with the magnitude of rising maize prices in recent years (Figure 1), and the impact that the increase in maize prices should have on poor, net-consumers of maize. In short, the initial findings from GOM call for a rigorous investigation that directly measures the causal impacts of the FISP program on poverty in Malawi. Several household-level studies provide useful complementary analysis that generally confirm and build upon the conclusions from the GOM s descriptive analysis of poverty in Malawi. Ricker- Gilbert and Jayne (2011) assessed the impact of the FISP on income and asset wealth using data from 2006/07 and 2008/09. They found that an additional kg of subsidized fertilizer acquired by a farmer in the current year boosted net crop income by US$1.16, on average. This benefit can be contrasted with the full retail price of fertilizer in Malawi, which ranged from roughly US$0.55 to $0.90 per kg during the panel period, although recipient farmers paid only US$0.10 to $0.15 per kg. The study also found that the subsidy program made no significant contribution to the average household s asset wealth over time. In a related study, Chirwa (2010) used average treatment effects and found for 2005/06 that, compared with households that did not receive FISP coupons, households that received the full FISP package consisting of two fertilizer coupons and one maize seed coupon had annual expenditure per capita that was US$11.19 or 8.2% higher. The two household-level studies mentioned above provide some evidence of statistically significant positive impacts of FISP on household Centre for Agric. Res. & Dev, Malawi 36

43 well-being. The economic magnitudes of these impacts are not huge, but they indicate that recipients of agricultural input subsidies do benefit from acquiring the productive inputs at reduced prices. While beneficiary households may have experienced some income gains, the findings from these studies could still be consistent with the GOM s findings of virtually no reductions in poverty for the following reasons. First, only 50%-70% of households acquired FISP coupons in any given year. Second, as mentioned earlier, there is evidence that more subsidized maize seed and fertilizer went to relatively better-off households. This would likely hinder the program s ability to reduce poverty. Third, in another study, Ricker-Gilbert and Jayne (2012) used quantile regression on household-level data in Malawi, and found that households in the top 10 th percentile of the crop income distribution received a US$1.23 increase in income from a kg of subsidized fertilizer. Conversely, households in the bottom 10 th percentile of the crop income distribution did not have any statistically significant boost in income from acquiring subsidized fertilizer. This result may indicate that the poorest households were the least able to use fertilizer effectively, due to limited landholding, poor soil quality, and lack of management knowledge. Therefore, if better-off households gained substantially more than poorer households when they participated in the program, the finding in Ricker-Gilbert and Jayne (2012) would be consistent with the NSO s result that income inequality has increased substantially in Malawi over the last seven years. 5. Conclusions and Implications for Research and Policy The recent re-introduction of agricultural input subsidy programs in a number of African countries has received a mixed reaction. In the case of Malawi, its Farm Input Subsidy Program (FISP) has been criticized for large, unsustainable costs by some observers, while others have hailed it as a success story. This article reviewed farm-level impact assessment studies of the Malawi FISP, in order to identify key gaps in the current literature, provide suggested avenues for future research, and recommend policies for improving the program. Malawi provides a useful case study, due to the country s long-term commitment to agricultural input subsidies, and because it is a learning case for other countries in the region who are considering agricultural input subsidy programs. The current objectives of the Malawi FISP are to increase maize production, promote food security, and enhance rural incomes by targeting subsidized input coupons to the productive poor. Our review of previous studies evaluating the FISP raises a number of questions on the appropriateness and effectiveness of using the FISP to reach these objectives, but further research is needed to reach sound conclusions. While national production estimates suggested dramatic increases in maize production and productivity (Figure 1), farm-level studies found relatively modest increases in maize production and yields under FISP (Ricker-Gilbert and Jayne, 2011; Chibwana et al., 2010; Holden and Lunduka 2010b). Consistent with the farm-level results of modest maize production increases under FISP, the country has imported maize during many of the FISP years until recently (FEWSNET, 2013), which raises questions on the program s ability to promote food security. There is a need for additional research at farm and national levels to determine the program s effectiveness in increasing maize productivity and production. Such an evaluation should include research that assesses the ability of the rural poor to effectively use modern inputs. Targeting the poor with subsidized input coupons, as FISP aims to do, may in fact preclude increasing maize productivity if poor farmers are unable to use modern seed and fertilizer effectively, because they lack complementary resources of labor, land, and managerial skills. There is some evidence in support of this hypothesis (Ricker-Gilbert and Jayne, 2012). A related and highly important area of research would investigate why the observed increase in maize production under the FISP was not accompanied by a reduction in maize prices (Figure 1). One recent study found that the FISP had a very modest effect at decreasing maize prices (Ricker-Gilbert et al., this issue). The FISP s impact on maize prices has major welfare implications for the majority of Malawi s smallholder farmers that are net consumers of maize and for poor urban consumers who spend a significant share of their income on maize purchases. Therefore more research should be conducted to inform this issue. Achieving national and household food security is among the main objectives of FISP. Thus, research should directly measure the effects of the program on these outcomes, but our literature review found no such studies. Regarding the program s objective of enhancing rural incomes, farm-level studies showed beneficiary households experienced some modest income gains (Ricker-Gilbert and Jayne, 2011; Chirwa, 2010), but a study by the GOM (2012b) found Malawi s poverty incidence was essentially the same before and during FISP, while income inequality increased substantially over the period. The program s ability to reduce poverty has likely been hindered because better-off households received more subsidized maize seed and fertilizer (Chibwana et al., 2012a; Holden and Lunduka, 2012a; Ricker-Gilbert et al., 2011). Furthermore, there is evidence that better-off households gained substantially more than poorer households when they participated in the program (Ricker-Gilbert and Jayne, 2012). Together these findings cast some doubt on FISP s ability to reduce poverty and call for Centre for Agric. Res. & Dev, Malawi 37

44 a rigorous investigation that directly measures the causal impacts of the FISP program on poverty in Malawi. Finally, given the high cost of the program and evidence of positive but modest farm-level impacts on maize productivity and economic wellbeing, there is a need for research to understand whether other possible investments would be more productive and efficient than FISP. These other possible investments include conditional cash transfers, food for work programs, or investments in public goods like road or market information services. The benefit-cost ratios of the program from 2005/06 to 2008/09 showed that the country has spent more implementing the program than it has benefited in terms of the value of increased maize output (Table 3). For example, results from Table 3 indicate that in most years the returns from subsidized fertilizer were nearly equivalent or lower than giving the recipient households cash. Political imperatives in Malawi, and many other African countries, indicate that agricultural input subsidies will continue for the foreseeable future, despite their high fiscal burden and uncertain effectiveness at increasing crop productivity, enhancing rural incomes, and promoting food security. It is thus important to consider the implications of existing research for how future rounds of the FISP can more effectively achieve its objectives. Under the previous presidential administration, the FISP was highly political. Allowing political objectives to dominate the design and implementation of the program made it difficult to introduce technical improvements to the FISP that might have allowed it to achieve its stated goals. Under the new administration, Malawi has the opportunity to de-politicize the program, make improvements, and re-focus it towards areas and people who can benefit the most from it. One obvious, though not necessarily easy, way to improve the FISP s effectiveness is the timely delivery of inputs. As mentioned earlier, there reportedly has been improvement in this area in recent years and it is important that the timely delivery of inputs continue. This is crucial to ensure farmers are able to use the subsidized inputs effectively, given research suggesting the maize response rate is significantly higher when fertilizer is applied at an appropriate time in the growing cycle (Holden and Lunduka 2010b). For subsidized inputs to be delivered in a timely manner, the Malawi government will likely need to start implementing the FISP soon after their budget session in July. Logistic activities like identification of households, or issuing of contracts to supply and transport fertilizer would need to be hastened to allow delivery of fertilizer before the growing season. Increased private sector involvement in the subsidy program can also ensure that input markets remain viable. This is highly important in case the FISP has to be scaled down or removed. Understanding how the private sector has been impacted over the years of FISP s implementation is also an important research question that needs to be addressed. Reviewed research suggested there are both pros and cons to targeting the program to the asset poor. Targeting of poor households was found to minimize crowding out effects (Ricker-Gilbert et al., 2011) and would likely reduce the gender gap in the adoption of modern maize (Fisher and Kandiwa, 2013), but research indicated such targeting may be ineffective at increasing maize productivity (Ricker- Gilbert and Jayne, 2012). Policy makers must consider these trade-offs when determining the best approach to distributing subsidized input coupons to achieve program objectives. Other factors that should be considered are the extra administrative costs of a targeted coupon system, the challenges of targeting certain groups in the Malawi context due to cultural norms, and evidence that targeting has not worked well in some years of the program. The extra administrative costs of a targeted system might be better spent directly on the subsidized inputs. A general (i.e. untargeted) subsidy for every resident rural household, based on a publicly displayed list of resident households for open validation in every village, and based on the recent population census, appears an attractive alternative. Since initial research suggested the FISP has been associated with little, if any, poverty reduction, additional programs will be needed that are specifically aimed at assisting the poor, for example agricultural extension activities to improve farm management knowledge and enable the efficient use of subsidized inputs, and safety-net programs, such as a cash transfer system, to boost income and wealth. References Adger, W. N., Huq, S., Brown, K., Conway, D., Hulme, M., Adaptation to climate change in the developing world. Progress in Development Studies 3(3), Awudu, A Spatial and Vertical Price Transmission in Food Staples Market Chains in Eastern and Southern Africa: What is the Evidence? Paper prepared for the Conference on Staple Food Trade and Market Policy Options for Promoting Development in Eastern and Southern Africa, 1-2 March Rome, Italy. Burke, W.J Maize Production in Zambia and Regional Marketing: Input Productivity and Output Price Transmission. Ph.D. dissertation, Michigan State University. Chibwana, C., Fisher, M., Jumbe, C., Masters, W., Shively, G., Measuring the impacts of Malawi s Farm Input Subsidy Program. Paper presented at the 2010 meeting of the African Agricultural Economics Association, September 20-23, Centre for Agric. Res. & Dev, Malawi 38

45 Chibwana, C., Fisher, M., Shively, G., 2012a. Cropland allocation effects of agricultural input subsidies in Malawi. World Development 40(1), Chibwana, C., Jumbe, C., Shively, G., 2012b. Agricultural subsidies and forest clearing in Malawi. Environmental Conservation 40(1), Chinsinga, B., The political economy of agricultural policy processes in Malawi: A case study of the Fertilizer Subsidy Program. Paper presented at the Futures Agriculture Conference on the Political Economy of Agricultural Policy in Africa. March 18-19, Pretoria, South Africa. Chirwa T.G., Program evaluation of agricultural input subsidies in Malawi using treatment effects: methods and practicability based on propensity scores. Munich Personal RePEc Archive (MPRA). Accessed March 2013, available at Chirwa, E.W., Mvula, P.M., Dorward, A., Matita, M., Gender and intra-household use of fertilizers in the Malawi farm input subsidy program. Future Agriculture Working Paper 02. Accessed March 2013, available at eagriculture/fac_working_paper_028.pdf. Denning, G., Kabambe, P., Sanchez, P., Malik, A., Flor, R., Harawa, R., Nkhoma, P., Zamba, C., Banda, C., Magombo, C., Keating, M., Wangila, J., Sachs, J., Input subsidies to improve smallholder maize productivity in Malawi: toward an African green revolution. Plos Biology 7 (1), Di Falco, S., Bezabih, M., Yesuf, M., Seeds for livelihood: Crop biodiversity and food production in Ethiopia. Ecological Economics 69(8), Dorward, A., Chirwa, E., 2012 Evaluation of the 2011/12 farm input subsidy program, Malawi report on programme implementation and benefit cost analysis. Report on programme implementation. School of Oriental and African Studies, London. Dorward, A., Chirwa, E., The Malawi agricultural input subsidy programme: to International Journal of Agricultural Sustainability 9(1), Dorward, A., Chirwa, E., Kelly, V., Jayne, T. S., Slater, R., Boughton, D., Evaluation of the 2006/07 agricultural input subsidy programme, Malawi. Final Report. Lilongwe, Malawi. Dorward, A.R., Chirwa, E., Slater, R., Evaluation of the 2008/9 agricultural input subsidy programme, Malawi: Report on programme implementation. School of Oriental and African Studies, London. Dugger, C. W., Ending famine, simply by ignoring the experts. New York Times, December 2, 2007, pp. 1. Fan, S., Johnson, M., Saurkar, A., Makombe, T., Investing in African agriculture to halve poverty by ReSAKSS Working Paper No. 25. International Food Policy Research Institute (IFPRI), Washington, D.C. FEWNETS, Malawi food security update. USAID-MALAWI. Available at Fisher, M., Kandiwa, V., Can agricultural input subsidies reduce the gender gap in modern maize adoption? Evidence from Malawi. Manuscript submitted for publication. Fisher, M., Shively, G.E., Agricultural subsidies and forest pressure in Malawi's Miombo woodlands. Journal of Agricultural and Resource Economics 32(2), Government of Malawi and World Bank, Malawi poverty and vulnerability assessment: investing in our future, Washington, D.C., USA: The World Bank and Government of Malawi. Government of Malawi, Malawi records yet another maize surplus. Which areas will be food insecure? The Malawi Vulnerability Assessment Committee Bulletin 7(1), Vulnerability Forecast, April 2011 to March Government of Malawi, 2012b. Integrated Household Survey Household Socio-Economic Characteristics Report. National Statistical Office, Zomba. Harrigan, J., Food insecurity, poverty and the Malawian starter pack: Fresh start or false start? Food Policy 33(3), Holden, S. T. and Lunduka, R.W., forthcoming, Input Subsidies, Cash constraints and Timing of Input Supply. American Journal of Agricultural Economics. Holden, S. T., Lunduka, R.W., 2012a. Who benefit from Malawi s targeted farm input subsidy program? Forum for Development Studies 39(3), Available at 8. Holden, S. T., Lunduka, R., 2012b. Do Fertilizer subsidies crowd out organic manures? The case of Malawi. Agricultural Economics 43(3), Holden, S.T., Lunduka, R., 2010a. Impacts of the fertilizer subsidy program in Malawi: Targeting, household perceptions and preferences. Department of Economics and Resource Management. Norwegian University of Life Sciences. April Holden, S.T., Lunduka, R., 2010b. Too poor to be efficient? Impacts of the targeted fertilizer subsidy program in Malawi on farm plot level Centre for Agric. Res. & Dev, Malawi 39

46 input use, crop choice and land productivity. Department of Economics and Resource Management. Norwegian University of Life Sciences. May Jayne, T.S., Chapoto, A., Minde, I., Donovan, C., The 2008/09 food price and food security situation in eastern and southern Africa: Implications for immediate and longer run responses. MSU International Development Working Paper No. 97. East Lansing, Michigan. Jayne, T. S., Sitko, N., Ricker-Gilbert, J., Mangisoni, J., Malawi s Maize Marketing System. The Evaluation of the 2008/9 Agricultural Input Subsidy Programme, Malawi, funded by DFID. Kelly, V.A., Boughton, D., Lenski, N., Malawi Agricultural Inputs Subsidy Program: Evaluation of the 2007/08 and 2008/09 Input Supply Sector Analysis. Levy, S., ed Starter Packs: A Strategy to Fight Hunger in Developing Countries? Lessons from the Malawi Experience, Wallingford, Oxfordshire, U.K.: CABI Publishing. Levy, S., Barahona, C., Findings of the Starter Pack and TIP M&E Programmes: Implications for Policy in and Beyond. London: DFID. Logistic Unit, Final Report. Implementation of Agricultural Input Subsidy Programme 2008/09. Logistic Unit, April Mason, N., Jayne, T., forthcoming. Fertilizer subsidies & smallholder commercial fertilizer purchases: Crowding out, leakage, & policy implications for Zambia. Journal of Agricultural Economics. Mason, N., Ricker-Gilbert, J., Disrupting demand for commercial seed: Input subsidies in Malawi and Zambia. World Development 45:75-91 Minde, I., Jayne, T., Crawford, E., Ariga J., Govereh, J., Promoting fertilizer use in Africa: Current issues and empirical evidence from Malawi, Zambia, and Kenya. Working Paper for Re-SAKSS/Southern Africa. Pretoria, South Africa. Ministry of Agriculture and Food Security, The 2007/2008 Input Subsidy Programme Review Report. An Internal Review. MoAFS, Lilongwe, Malawi. Minten, B., Barrett, C., Agricultural technology, productivity, and poverty in Madagascar. World Development 36(5), Myers, R Evaluating the Efficiency of Inter- Regional Trade and Storage in Malawi Maize Markets. Report for the World Bank. East Lansing, MI: Michigan State University. Myers, R.J., Jayne, T.S., Multiple-Regime Spatial Price Transmission with an Application to Maize Markets in Southern Africa. American Journal of Agricultural Economics 94: Norwegian Agency for Development Cooperation, Common Approach to Budget Support (CABS) in Malawi, NORAD collected reviews, Oslo ISBN Pan, L., Christiaensen, L., Who is vouching for the input voucher? Decentralized targeting and elite capture in Tanzania. World Development. 40(8): Peterman, A., Behrman, J., Quisumbing, A. R., A review of empirical evidence on gender differences in non-land agricultural inputs, technology, and services in developing countries. IFPRI Discussion Paper Poverty, Health, and Nutrition Division (PHND). International Food Policy Research Institute. Washington, D.C. Ricker-Gilbert, J., Jayne, T., Shively, G., Addressing the Wicked Problem of Input Subsidy Programs in Africa. Applied Economics Perspectives and Policy, forthcoming. Ricker-Gilbert, J., Jayne, T.S., What are the enduring effects of fertilizer subsidies on recipient households? Staff Paper Department of Agricultural Food and Resource Economics, Michigan State University. East Lansing, MI. USA. Ricker-Gilbert, J., Jayne, T.S., Chirwa, E., Subsidies and crowding out: a double-hurdle model of fertilizer demand in Malawi. American Journal of Agricultural Economics 93(1), Ricker-Gilbert, J., Jayne, T.S., Do fertilizer subsidies boost staple crop production and reduce poverty across the distribution of smallholders in Africa? Quantile Regression Results from Malawi. Accessed March 2013, available at Paper16263.html. Ricker-Gilbert J., Mason, N.M., Jayne, T.S., Darko, F.A., Tembo, S., General equilibrium effects of input subsidy program on maize prices: Evidence from Malawi and Zambia. Agricultural Economics (this issue). Sauer, J., Tchale, H., The economics of soil fertility management in Malawi Appl. Econ. Perspect. Pol. 31 (3): Snapp, S.S., Rohrbach, D. D., Simtowe, F., Freeman, H. A., Sustainable soil management options for Malawi: can smallholder farmers grow more legumes? Agriculture, Ecosystems, and Environment 91(1 3), Tostau, E., Brorsen, W., Spatial Price Efficiency in Mozambique s Post-Reform Maize Markets. Agricultural Economics 33: Centre for Agric. Res. & Dev, Malawi 40

47 Xu, Z., Burke, W. J., Jayne, T. S., Govereh, J., Do input subsidy programs crowd in or crowd out commercial market development? Modeling fertilizer demand in a two channel marketing system. Agricultural Economics 40, Centre for Agric. Res. & Dev, Malawi 41

48 Table 5: Empirical studies of farm-level impacts of the Malawi FISP Study Geographical focus Dataset Information Outcome(s) measured Empirical approach Chibwana et al. Kasungu and Maize yields (2010) Chibwana et al. (2012a) * Chibwana et al. (2012b) * Chirwa (2010) Chirwa et al. (2011) Fisher Kandiwa (2013) Holden Lunduka and and Machinga Districts Kasungu Machinga Districts and Chimaliro and Liwonde Forest Reserves in Kasungu and Machinga Districts Malawi (nationally representative) Malawi (nationally representative) Malawi (nationally representative) Kasungu, Lilongwe, 380 hhlds. in 2008/09 (cross section) 380 hhlds. in 2008/09 (cross section) 380 hhlds. in 2008/09 (cross section) 1,147 hhlds in 2005/06 (cross section) 4,727 maize plots in 2008/09 (cross section) 11,221 maize plots in 2009/10 (cross section) 450 hhlds. in 2005/06, FISP coupon receipt and cropland allocation Forest clearing Annual household real expenditure per capita Probability of commercial and/or subsidized fertilizer Monetary value of the FISP coupon redeemed and modern maize adoption by male and female farmers Maize share, area tree Instrumental variables estimation Instrumental variables estimation Instrumental variables estimation Propensity score matching and instrumental variables estimation Probit regression Instrumental variables estimation and simulations based on the logit adoption model Household fixed-effects Key finding(s) Maize yield was positively associated with receipt of subsidized fertilizer. Asset-poor and female-headed households were less likely to receive FISP coupons; receipt of a FISP coupon was associated with an increase in the share of land planted in maize. Households that received FISP coupons had lower forest clearing than those that did not receive; indirect negative forest effects arose due to the use of trees to construct tobacco drying sheds. Hhlds that received the full package of fertilizer coupons had annual expenditure per capita that was 8.2% or US$13 higher than nonrecipient hhlds. There was no significant difference in fertilizer application by gender of the plot manager for hhlds that received subsidized fertilizer and did not have access to other fertilizer. Female householders had a higher value of redeemed coupons than male heads; the FISP did not seem to influence the gender gap in modern maize adoption but simulations suggest it could reduce the gap substantially if the poorest households were targeted. Better access to input subsidies over time Centre for Agric. Res. & Dev, Malawi 42

49 (2010b) Holden Lunduka (2012a) * Holden Lunduka (2012b)* and and Mason and Ricker-Gilbert (forthcoming)* Ricker-Gilbert and Jayne (2011) Ricker-Gilbert and Jayne (2012) Ricker-Gilbert et al. (2011)* Machinga, Zomba, Chiradzulu, and Thyolo Districts Kasungu, Lilongwe, Machinga, Zomba, Chiradzulu, and Thyolo Districts Kasungu, Lilongwe, Machinga, Zomba, Chiradzulu, and Thyolo Districts Malawi and Zambia (nationally representative) Malawi (nationally representative) Malawi (nationally representative) Malawi (nationally representative) 2006/07, 2008/09 (panel) 450 hhlds. in 2005/06, 2006/07, 2008/09 (panel) 450 hhlds. and 3,000 farm plots in 2005/06, 2006/07, 2008/09 (panel) 1,375 households in 2006/07, 2008/09 (panel) in Malawi 1,375 households in 2006/07, 2008/09 (panel) 2,968 households in 2003/04, 2006/07, 2008/09 (panel) 2,406 households in 2003/04, 2006/07 (panel) planting, and forest clearing FISP coupon receipt, maize production, and food security Probability and intensity of organic manure use Crowding out of commercial seed by subsidized seed and fertilizer Maize production, net-value of crop production, income, asset wealth Maize production and the value of crop output Farmer demand for commercial fertilizer regression analysis Ordered probit models and ordinary least squares (OLS) Correlated random effects combined with a control function approach Correlated random effects estimation with control function method First Difference estimation with control function Quantile regressions with correlated random effects with control function Correlated random effects combined with a control function was associated with maize intensification and smaller maize area planted. The FISP did not target the poor; the program enhanced maize production and food security. Access to subsidized inputs was found to crowd out manure use, but only to a small extent. On average, one additional kg of subsidized seed crowded out 0.56 kg of commercial seed. Small positive effect from subsidized fertilizer on maize production in current year and over time. No significant effect on asset wealth from the subsidy. Returns to subsidized fertilizer were much higher for those at the top of the maize production distribution. On average, one additional kg of subsidized fertilizer crowded out 0.22 kg of commercial fertilizer. * indicates a peer-reviewed journal publication. Centre for Agric. Res. & Dev, Malawi 43

50 Table 6: Important features of the FISP, 2005/06 to 2010/11 Cropping year 2005/ / / / / /11 Total fertilizer subsidized (MT) 131, , , , , Total maize seed subsidized (MT) n/a 4,524 5,541 5,365 8,652 8,000 Total legume seed subsidized (MT) n/a 1,551 1,600 Redemption price (US$/50kg bag) Fertilizer subsidy paid by government (%) Total program cost (US$ million) Total cost as % agricultural budget n/a Total cost as % of national budget Source: Logistics Units reports (2009); MoAFS (2008); Dorward and Chirwa (2011) Centre for Agric. Res. & Dev, Malawi 44

51 Table 7: Benefit-cost ratios of subsidized fertilizer, 2005/06 to 2008/09 Data sources: Author calculations based on data from Dorward and Chirwa (2011) and Ricker- Gilbert and Year (A) Commercial price of maize (US$/kg) (B) Government cost per kg of fertilizer (US$/kg) (C) Subsidized price of fertilizer (US$/kg) B/C ratio Unconditional maize response rate of 5kg maize / kg fertilizer* (D) At government cost per kg of fertilizer Conditional mean maize response rates of 2.71kg maize /kg fertilizer (5*A)/B (2.71*A)/C (2.71*A)/B 2005/ / / / Jayne (2012). *The figures in Column D above come from assumed mid-maize grain response rate of 15kg grain per kg nitrogen in Dorward and Chirwa (2011). (E) At government cost per kg of fertilizer (F) At subsidized fertilizer prices Centre for Agric. Res. & Dev, Malawi 45

52 Table 8: Informal net cross-border maize imports to Malawi, 2004/05 to 2012/13 marketing years Marketing Year Mozambique Tanzania Zambia 2004/05* 71,229 2,019 2,123 75, /06 71,085 83, , /07 76,803-1, , /08 52, , /09 54,020 2,671 5,259 61, /10 54,275-5, , /11-3,652-6, , /12-30,651-17, ,866 total cross border net imports 2012/13** 10,429-24, ,912 average net imports per year (2005/06 to 2011/12) 39,172 7, ,867 Note: Marketing year runs from April to March in Malawi; (Net Imports= Total Imports-Total Exports); * data for 2004/05 cover July-March only; **data for 2012/13 cover April-January only; Source FEWSNET Figure 6: Maize production and prices, Malawi 1991 to FISP years Maize price (US$/MT) year Maize production ('000 MT) Maize price (US$/MT) 0 Data source: FAOSTATS. Centre for Agric. Res. & Dev, Malawi 46

53 Annex 1.3: The Role of Private Input Suppliers in Smallholder Agriculture: Insights from Malawi By Stevier Kaiyatsa, Department of Agricultural and Applied Economics, Lilongwe University of Agriculture and Natural Resources (LUANAR), Bunda College of Agriculture, P. O. Box 219, Lilongwe, Malawi 1. Introduction Increased use of advanced agricultural technologies such as inorganic fertilizers and improved seeds to achieve substantial agricultural productivity growth in Sub-Saharan African (SSA) countries is widely recognized. According to Kelly et al. (2003) promotion of adoption of improved agricultural technologies is aimed at meeting economic growth, poverty reduction and food security goals. For instance, in the period from to overall fertilizer use and use per hectare rose by 16% and 5% respectively. However, Seini et al. (2011) indicated that African agriculture has not performed well as expected resulting in declining per capita food production over the past decades. This has been due to limited access to credit by smallholder farmers, high cost and unavailability of inputs, inefficient agricultural input markets, and the absence of a conducive policy environment. Despite low uptake of modern agricultural technologies by smallholder farmers, Kelly et al. (2003) noted that there has been significant progress in developing input markets in some SSA countries. To promote supply response, African governments have participated in the delivery of either domestically or outsourced modern agricultural inputs, often at subsidized prices, free of charge or on credit to smallholder farmers. On the other hand, modern input has been provided by private input suppliers at full market prices. Seini et al. (2011) indicated that in most SSA countries, the agricultural input markets are not well organized and suggested that there is need for a well-organized and transparent input market system and that transparency must transcend from input imports or manufacturing through wholesalers and agro-dealers down to the smallholder farmers. Input supply sector in Malawi comprise of Government and private input suppliers. Kelly et al. (2010) indicated that the participation of the Government in the input market (at both procurement and distribution level) tends to vary from year to year, depending on decisions about input support programs. The inputs are distributed through SFFRFM depots and ADMARC market units across the country. Since independence, the Malawi Government has undertaken a lot of initiatives to intervene in the input market in order to increase smallholder farmers farm productivity and food security. For instance, according to World Bank (2005), in the 1980s, fertilizer use was supported by a credit program that provided a universal fertilizer subsidy to primarily wealthier smallholder farmers which was followed by free distribution of seed and fertilizer in subsequent years. The 1997/98 food crisis led to implementation of the Universal Starter Pack because very few smallholder farmers were able to purchase small amounts of the best bet 14 maize productivity packages. As of 2005/06 growing season, the Malawi Government started implementation of a large scale agricultural input subsidy programme (Dorward and Chirwa, 2011). Following adoption of liberalization policies, Imperial College et al., (2007) pointed out that there has been increased participation of private sector in both fertilizer and seed markets in Malawi. The private sector comprises both large traders such as Farmer s World and smaller scale agro-dealers. The private sector participates from the procurement to the retail level. The players have organized themselves into professional associations in order to improve professionalism in the sector and to represent their members in policy discussions with the Government of Malawi (Imperial College et al., 2007; Kelly et al. 2010). Problem statement Kelly et al. (2010) indicated that one of the objectives of the Agricultural Input Subsidy Program is to build a reliable input distribution system with an appropriate mix of Government and private sector services. As such the Programme targets approximately 50 percent of smallholder farmers annually (Dorward and Chirwa, 2011) and leave out other farmers to access commercial inputs. Despite such Government interventions and more recently concerted efforts by Non-Governmental Organizations (NGOs) in promoting use of improved seeds and fertilizers, smallholder farmers use of improved agricultural technologies remain low (Simtowe and Zeller, 2006; GoM, 2012; Chirwa, 2005; Dorward and Chirwa, 2011). Low use of improved seed and fertilizer has been attributed to low emergency of the private sector (Morris et al., 2007) and the effect of input subsidies on crowding out commercial input purchases due to poor targeting of households that would otherwise buy the inputs at market prices (Mason and Ricker-Gilbert, 2012; Chirwa et al., 2010; Takeshima et al., 2012; Ricker- Gilbert and Jayne, 2009; Xu et al., 2008). 14 Technologies that are deemed to have particular value for identified farming environments or groups Centre for Agric. Res. & Dev, Malawi 47

54 Furthermore, the situation is exacerbated by high transport and transactions costs which lead to high input costs and in turn inhibit the development of input supply systems in less accessible areas (Imperial College et al., 2007). Justification Having an effective and sustainable private input supply sector is vital for development of the smallholder agriculture in Malawi. Agricultural input subsidies are widely used as policy instruments to develop a vibrant private sector for the supply of inputs as well as raising farmers income and agricultural productivity (Takeshima et al., 2012). Thus, subsidies used for this purposes help to develop the sector to grow sustainably so that there will be no need for subsidies in the longer term. This can crowd-in the private input suppliers and boost demand for the inputs, thereby helping input dealers handle larger volumes and raise their profitability through economies of scale. Morris et al. (2007) indicated that sustained high profitability encourage suppliers to offer more inputs at the prevailing market prices which in turn draws new players into the market, thereby increasing supplies. As the business of private input suppliers grow, there is a high chance that their operations can be extended to the rural areas in order to increase their market share which in turn will smoothen out smallholder farmers access to commercial inputs. Objectives Overall Objective To assess the supply side effects of input subsidy program on the role of private input suppliers in smallholder agriculture Specific Objectives To identify factors that influence smallholder farmers decision to access the private input market. To determine the effect of farmer s proximity to market on their decision to access the private input market. To estimate the effect of input subsidy program on private input suppliers. To quantify the effect of proximity to ADMARC or other government depots distributing subsidized input on individual input suppliers businesses. Hypotheses Farmer characteristics have no effect on their decision to access the private input market. Farmer proximity to market has no effect on their decision to access the private input market. The input subsidy program has had no effect on private input suppliers. Proximity to ADMARC or other government depots that distribute subsidized input has no effect on individual input suppliers businesses. Research Questions Do socio-economic and institutional factors influence smallholder farmers decision to access the private input market? What is the effect of farmer s proximity to market on their decision to access the private input market? What is the effect of input subsidy program on private input suppliers? Does proximity to ADMARC or other government depots distributing subsidized input has an effect on individual input suppliers businesses? Literature Review Fertilizer industry in Malawi One of the noticeable changes in the fertilizer industry for the past 25 years is increased involvement of the private sector (Imperial College et al., 2007). Up to the early 1990 s only ADMARC was permitted to sell fertilizer (at a subsidy) to smallholder farmers and private importers were supplying ADMARC and commercial estates. With liberalization of the sector and the removal of the subsidy in the mid 1990s, the market opened up, and the private sector became increasingly involved. Imperial College et al. (2007) report that firms are particularly involved in the importation and supply of fertilizer to a network of over 400 retail outlets in Malawi (public and private sector combined) plus an informal network of independent agro-dealers. The business enterprises have organized themselves into active associations (Kelly et al. 2010; Imperial College et al., 2007). According to Imperial College et al. (2007) one of the associations is fertilizer importers association which act as a liaison between the fertilizer sector and the GoM in discussions of fertilizer policy. The other noticeable associations are CNFA/RUMARK and AISAM networks of agrodealers. Both networks were created from donor funded projects to act as a means of building a more vibrant private sector supply system in rural areas (Kelly et al. 2010; Imperial College et al., 2007). Another network is NASFAM which is a network of farmer cooperatives. Although these networks input supply activities are profit oriented, member cooperatives benefit from donor funding which Centre for Agric. Res. & Dev, Malawi 48

55 provides training and business management support. Government network is strong in terms of its ownership of 58 SFFRFM depots (Kelly et al report 56) and more than 600 ADMARC market units (Kelly et al report 666) across the country thereby serving as a longed-for input distribution or sales points for many farmers who are not served adequately by the private sector (Imperial College et al., 2007). With regard to the seed sector, Imperial College et al. (2007) report that there is an overlap between the seed sector and the fertilizer sector at the distribution and retail level. For less under-capitalized agrodealers, most retail outlets sell both seeds and fertilizers together. The Seed Trade Association of Malawi started operating in 2004 to enhance communication between the sector and the Ministry of Agriculture, promote their products and ensure that seeds sold on the market were of good quality. The CNFA and AISAM networks support agrodealers for both their fertilizer and seed trade. Among the services offered include business management and product training, credit guarantees (CNFA only) and periodic reports on market conditions. Performance of agricultural sector Agricultural productivity gains for staple food crops (e.g. maize) can be achieved when use of improved farm inputs such as inorganic fertilizers and improved seed is intensified. Dorward and Chirwa (2011) noted that widespread use of fertilizer on maize by smallholder farmers is constrained by the problem of profitability and affordability. In Malawi, from the mid-1990s to the mid-2000s unsubsidized fertilizer use was more profitable on maize produced for own consumption 15 than for sale. Affordability of fertilizers by poor farmers is a major problem during the cropping season as they faced with both hungry gap (when they need to invest labour, seed and other inputs in crop production, but also need to earn off farm income as food stocks form the previous season run out) and very high borrowing costs and a lack of low cost input finance services. Furthermore, household hungry gap problems are worsen off by depressed wage rates and asset prices and high food prices in the rural economy. Dorward and Chirwa (2011) pointed out that improving the profitability of fertilizer use in maize production requires lower fertilizer prices (emanating from either of greater efficiency in fertilizer supply and lower importation and distribution transport costs or of a subsidy), higher maize prices and/or greater efficiency in the use of fertilizer. High maize prices is 15 Due to farmers fears of the effects of a bad year on maize purchase prices difficult to achieve because 10% of Malawian maize producers are net sellers of maize and 60% are net buyers of maize; hence high maize prices disrupts people s livelihoods and food security. Because changes to maize prices and improved efficiency of fertilizer use will not increase the affordability of fertilizer for the majority of poor rural households, this resulted in the emergence of Farm Input Subsidy Programme in Malawi. Farm Input Subsidy Programme The core objective of the FISP has been to increase resource poor smallholder farmers access to improved agricultural inputs with the purpose of improving food self-sufficiency and to increase farmers incomes through increased food and cash crop production. The programme targets both improved seed and inorganic fertilizers although the targets keep on changing from time to time. According to Dorward and Chirwa (2011), from the period 2005/06 to 2009/10, FISP targeted maize and tobacco fertilizers. However, in the 2009/10 tobacco fertilizers were not subsidized. During the period the percentage subsidized price was over 60% of the market price. In addition, within the same period improved maize seed, cotton seed (2007/08 and 2008/09), legume seed (2007/08 and 2009/10), and cotton chemical (mainly in 2007/08) were also considered. The volumes of the subsidized inputs increased steadily from 2005/06 (108,986mt) to 2008/09 (182,309mt) and dropped in 2009/10 (161,495mt). Dorward and Chirwa (2011) indicate that distribution of the subsidized inputs was subjected to fraud, although is it difficult to determine due to inexistence of transparent and formal audit systems for the whole programme and to discrepancies between NSO and MoAFS estimates of the total farm families in Malawi. MoAFS estimates are 30% or more above NSO estimates. Collectively, lack of transparent and formal audit system and discrepancies between NSO and MoAFS resulted in allocation of vouchers to non-existent ( ghost ) beneficiaries (or villages), diversion to others (government staff, traditional leaders or politicians), direct allocation to people who did not satisfy beneficiary criteria, and printing of extra or counterfeit vouchers. Agricultural productivity gains arise from input use, proper agronomic practices (timely planting, weeding and fertilizer application), and favourable weather conditions (rainfall). Although it is difficult to obtain precise measurement of crop response to incremental input use (Dorward and Chirwa, 2011) substantial incremental production were recorded in the period from 2005/ /10. National crop production estimates for maize rose from 975,262mt in 2005/06 to 2,031,816mt in 2008/09. The crop estimates were lower than those in the widely cited national crop Centre for Agric. Res. & Dev, Malawi 49

56 estimates for maize production and had much lower variation. The crop production estimates have been over-estimated following the inception of FISP. However, food security at both national and household level improved substantially during the 2005/ /10 implementation period of the FISP. The average estimated net imports were over 1,000mt compared with nearly 132,000mt before the programme. In terms of food accessibility, the mean prices prior to the subsidy programme were 0.18US$/kg which are lower than the mean of 0.25US$/kg for marketing seasons following subsidy implementation (Dorward and Chirwa, 2011). Involvement of private input suppliers in input distribution system The second objective of the FISP is to promote the participation of private sector in input distribution system. Chirwa and Dorward (2012) indicate that the private sector has played a key role in several aspects of the subsidy programme including procurement of fertilizer, transportation to various markets, retail sale of fertilizer and the production and sale of improved seeds. As noted by Chirwa and Dorward (2012) participation of the private sector in the input market is believed (1) to promote efficiency as the sector is less prone to the bureaucracy associated with state delivery of services, (2) as a strategy of developing the private market system especially in remote areas where the incentives for private sector investment in markets is weak, (3) allow the Government to use scarce resources on other activities, by reducing the cost of the subsidy to the government, (4) to reduce the displacement effects of the input subsidy programme and (5) to reduce transaction costs and costs of queuing. Participation of private sector in the subsidy programme particularly in fertilizer and seeds components has been different. In fertilizer component, the participation of private input suppliers has varies from time to time. In some cropping seasons, the private sector has been included and in other seasons excluded from retail sales of fertilizers whereas for the seed component participation has been consistent. However, the sector has remained an important partner in the procurement of fertilizers for the programme and commercial sales in various market outlets across the country (Chirwa and Dorward, 2012). According to Dorward and Chirwa (2011) the period from 2005/ /10 cropping seasons, there was increased dependency of the government on private sector fertilizer imports to supply parastatal fertilizer sales, with improved tender procedures. The number of private bidders increased from 24 in 2009/10 to 65 companies in 2011/12. Similarly, the number of successful awards of contracts also increased from 10 private companies in 2009/10 to 20 private companies in 2011/12 (Chirwa and Dorward, 2012). This implies that the programme has over time attracted new participants. In terms of the volume, Chirwa and Dorward (2012) report that the volume accounted by the private sector increased from 70% in 2007/08 to 95% in 2010/11 and dropped to 78% in 2011/12. Commercial fertilizer sales also increased between 2004/05 and 2005/06 seasons and dropped sharply in 2006/07. In 2008/09 there was an increase in the quantity of fertilizer for commercial sales due to a small reduction in subsidized fertilizer between 2007/08 and 2008/09. There was a decline in importation and availability of commercial fertilizer in 2009/10 16 season and an increase in 2010/11 season (in which tobacco fertilizer was excluded) whereas the subsidy levels remained unchanged. In 2011/12 season there was a drop in imports, subsidized fertilizers and available commercial fertilizers partially due to the collapse of tobacco prices in 2010/11 season which led many farmers abandon the crop in 2011/12 season. An overall increase in importation and availability of commercial fertilizers implies that subsidy programme might have stimulated fertilizer use. Chirwa and Dorward (2012) report that the participation of the private sector in the retail marketing of subsidized fertilizers has been the most difficult aspect in relation to the development of the private input markets across the country. The private sector only in 2006/07 and 2007/08 seasons was allowed to redeem fertilizer vouchers and its market share was about 22% of subsidy fertilizer sales (Chirwa and Dorward, 2012; Kelly et al. 2010). Dorward and Chirwa (2011) indicate that in 2006/07 and 2007/08 there was a limited number of larger retail chains selling subsidized fertilizer with a buyback scheme to reduce government fertilizer stock holding risks, and a premium in 2007/08 to stimulate private retail network development in more remote areas. In 2008/09 the private sector was excluded from subsidized fertilizer retail sales in 2008/09. During the period there were also changes in the extent and modalities of private sector involvement in seed sales, with all seed supplies from private seed supplies, and seed sales through a large variety of retail outlets including small agro-dealers, with variable seed pack sizes depending on seed costs. Dorward and Chirwa (2011) report that it was difficult to determine the extent to which the programme has contributed to the development of improved input supply systems because impacts were supposed to be considered separately for fertilizer importers, fertilizer retailers, seed suppliers, and seed 16 The decline is associated with a sharp decline in the price of burley tobacco (Chirwa and Dorward, 2012) Centre for Agric. Res. & Dev, Malawi 50

57 retailers (both small independent agro-dealers and retail outlets of larger companies). Overall the private sector fertilizer importers benefited from supplying increasing volumes of government subsidized sales, though they faced some difficulties from exposure to foreign exchange losses with delays in payments in Malawi Kwacha. The maize seed suppliers also benefited from significant growth in sales. The effect of the programme on retail outlets was twofold namely; for those participating in the programme and those excluded. For those excluded in the programme were negatively affected due to direct losses from displacement of commercial sales by subsidy sales. On the other hand, those participating in the programme benefited from sales of subsidized inputs. Indirectly, all retail outlets gained from general increases in demand if the programme stimulated wider income growth and eased liquidity constraints. Small agro-dealers are excluded from retail sales of subsidized fertilizers since the inception of the programme. However, larger companies with retail outlets were able to sell subsidized maize seed from 2006/07 onwards. Some larger companies with retail outlets were also able to sell subsidized fertilizers in 2006/07 and 2007/08, but were excluded in 2008/09. As a result, participating retail outlets reported significant increase in unsubsidized sales in 2006/07 whereas the private retailers that were excluded in 2008/09 reported falls in unsubsidized fertilizer sales. Demand side impacts of farm input subsidy program on commercial input purchases Takeshima et al. (2012) indicated that subsidies can be a second best policy for the development of the commercial input sector, if the use of subsidies on inputs can crowd-in the commercial sector by addressing key market failures. Thus subsidies used for such purpose help the private sector to develop sustainably so that there will be no need for subsidies in the longer term. Fertilizer subsidies can crowd-in the private fertilizer sector if the subsidies serve to sensitize farmers as to the benefits of the use of fertilizer on their crops and boost demand for the input, thereby helping private fertilizer dealers handle larger volumes of fertilizer and raise their profitability through economies of scale. However, fertilizer subsidies may crowd-out the private fertilizer sector if fertilizer demand is not price elastic due to rapidly decreasing marginal returns on its use or if subsidies are received by those who are already using fertilizer bought from private non-subsidized sources. Empirical evidence reveals that fertilizer subsidies results in crowding-out of private fertilizer sector. For instance, one ton of subsidized fertilizer reduces the demand for commercial fertilizer between 0.19 and 0.35 tons in Nigeria (Takeshima et al,. 2012), whereas in Malawi 1kg of subsidized fertilizer reduces the demand for commercial fertilizer by 0.2kg (Ricker-Gilbert and Jayne, 2009) and in Zambia one ton of subsidized fertilizer reduces the demand for commercial fertilizer by 0.12 ton (Xu et al., 2008). Supply side impacts of farm input subsidy program Few studies have attempted to address the supply side effects of the farm input subsidy program in Malawi. For instance, Kelly et al. (2010) and Chirwa and Dorward (2012) analyzed the supply side effects of the program on fertilizer and maize sub-sectors using qualitative methodologies. Therefore, this study builds on the previous work (Kelly et al., 2010; Chirwa and Dorward, 2012) by employing both quantitative and qualitative methodologies to unveil the supply side effects of the subsidy program in fertilizer sector. The study by Freeman and Kaguongo (2003) examined the factors influencing the entry and sales decision of private traders in fertilizer retail trade in a liberalized market in Kenya using a Heckman twostage econometric model. Freeman and Kaguongo (2003) postulated that a trader follows a two-stage process in making fertilizer trading decision. The first stage is an entry decision in which the trader decides whether or not to sell fertilizer. In the second stage, depending on the entry decision, the trader makes the fertilizer sales decision. The study will improve and extend on their work by providing an overall unconditional APE estimate of how much commercial fertilizer supplied is displaced by subsidy and test whether traders make fertilizer market participation decision sequentially or simultaneously. Methodology Conceptual framework Smallholder farmer s usage of improved farm inputs such as inorganic fertilizers and improved seed is either through subsidized fertilizers or private input suppliers at market price. By design FISP targets resource-poor Malawians who own a piece of land and are resident in the village, with special consideration to guardians looking after physically challenged persons and vulnerable groups such as child-headed households, female-headed or orphanheaded households and households affected by HIV and AIDS (Chirwa et al., 2010). Centre for Agric. Res. & Dev, Malawi 51

58 Figure 7: Factors affecting household access and supply of inorganic fertilizers by private input suppliers (Hypothesized) Institutional factors Subsidy fertilizer Socioeconomic factors Climatic events Private input suppliers Commercial Inorganic fertilizer usage Farm Productivity Finance On-farm investment Farm income Socio-economic and institutional factors influence farmer s usage of commercial inputs. Usage of improved inputs has potential of improving agricultural productivity which in turn increases farm income. If increased farm income is re-invested in agriculture, farmers can access improved farm inputs through the private input suppliers. Indirectly, increased access of farmers to inputs supplied by private traders can help to develop the input supply system in the economy because of increased profits. However, input supplier business is affected by FISP which may crowd-in or crowd-out private purchases. Furthermore, private traders are financially constrained which also negatively affect their performance in the input market. Freeman and Kaguongo (2003) indicate that participation in fertilizer markets is conditioned by traders willingness and capacity to invest. But the variables that influence willingness and capacity to invest may differ. Some traders may be willing to sell fertilizer but prevented from doing so because of various constraints. The willingness to participate in fertilizer trade is conditioned by asset or liquidity position, the state of physical infrastructure, skills or experiences in fertilizer trade, and ex-ante transaction costs of actually trading in fertilizer. On the other hand, the capacity to sell fertilizer is conditioned by relative returns to fertilizer trade and trader s risk attitudes. Proposed model specifications Factors influencing smallholder farmers decision to access the private input market Participation of smallholder farmers in fertilizer markets can boost farm productivity which eventually can be translated into improved household income and food consumption. Farmers can access farm improved inputs such as fertilizer through (a) private input suppliers, (b) fertilizer subsidy program, or (c) through a combination of private input suppliers and fertilizer subsidy program. Mabiso (2012) indicates that smallholder farmers are considered to be rational utility-maximizing decision makers, choosing out of a set of market participation opportunities to either participate in a fertilizer related market or not. This multinomial decision is based on the farm household s utility obtainable from participation subject to a reservation utility, resource constraints and farm household characteristics. It is assumed that a latent random utility model generates the observed multinomial participation variable. Hence, Multinomial Logit (MNL) model will be used to examine the factors that influence farmers choice in the face of different options because the dependent variable, sources of fertilizer, has more than two categories with no natural ordering. Smallholder farmer s choice of fertilizer source is formalized as follows: Suppose the unobserved variable is the farmer s utility if the farmers chooses fertilizer source. If each farmer chooses the optimal fertilizer source which brings the highest Centre for Agric. Res. & Dev, Malawi 52

59 utility level, the observed choice of the farmer source can be expressed as follows: for function Where Is a linear is household indicator while correspond to supply sources. is a vector of parameters to be estimated, is the error term. is a vector of explanatory variables. Under the assumption that error term are independent and identically distributed, the above probability function can be written as follows: Where, is the logistic distribution function. By implication, it is supposed that the Independence of irrelevant alternatives hypothesis is respected. The above function is estimated by maximizing the following Log Likelihood function: Where and 0 otherwise. Under certain conditions, the Maximum Likelihood method provides consistent and efficient estimates of the parameters (Folk and Sikod, 2012). Explanatory variables Chirwa et al. (2010) report that male-headed households, older household heads, famers with larger parcel of land, farmers that earn cash income from agriculture and households that are food secure are more likely to receive fertilizer coupons than their counterparts. Hence, it is expected that as maleheaded household, age of household head, farm size, crop income, and household food security status (see Table 1) increase, the likelihood of farmers fertilizer access through subsidy program relative to private input suppliers. On the other hand, Chirwa et al. (2010) observe that households that accessed commercial fertilizer in the previous season and existence of household members in labour market either through salaried or ganyu employment are less likely to access fertilizer coupons. It is, therefore, expected that the likelihood of these farmers to access fertilizer through private input suppliers will be higher compared to other sources. Table 9: Explanatory variables for MNL model Variable Labour Sex Age Educ Farm_size Crop_income Access_crdt Pricemz Food_gap Previous_commfrt Description Number of household members in labour market Gender of household head. 1=if the head is male, 0 otherwise Age of head in years Number of years the household head spent in school Farm size in hectares Crop revenue in (MK 000) Household access to credit (dummy variable) Historical maize prices (dummy variable) Number of food gap months Dummy variable equal to 1 if the household accessed commercial fertilizer in the previous season It is hypothesized that education level of household head, high previous maize prices, and access to credit are expected to have a positive influence on farmer s choice of fertilizer source from private input suppliers as compared to other sources. The effect of farmer s proximity to market on their decision to access the private input market The effect of farmer s proximity to market on their decision to access commercial inputs will be estimated by examining the quantities of fertilizers purchased from the private input suppliers. The dependent variable thus, has a censored distribution since there is a cluster of households (those that do Centre for Agric. Res. & Dev, Malawi 53

60 not access) with zero value; hence a Tobit model to correct for the non-normality of the distribution of the censored dependent variable will be used: Where is a latent variable representing amount of commercial fertilizer purchased (kg); is farmer s distance to private input market (km); Z is a vector of household level exogenous factors (household members in labour market, sex of head, age of head, farm size, crop income, food gap, previous access to commercial fertilizer, previous maize price, education level of head and access to credit), and are parameters to be estimated. The effect of farmers proximity to the input private market is captured by. The effect of input subsidy program on private input suppliers Private input supplier s fertilizer market participation and the supply of commercial fertilizer can be estimated using a corner solution model where a nontrivial portion of the data is at zero (Ricker-Gilbert and Jayne, 2009; Yu, 2007), although Heckman selection model may seem plausible. Heckman selection approach is applicable for incidental truncation where the zeros encountered in the data set are missing values. A corner solution model is more appropriate than a selection model for this problem because inorganic fertilizer has been available for decades in Malawi (Ricker-Gilbert and Jayne, 2009); hence it is assumed that the majority of agro-dealers are aware of fertilizer business. As such, the zeros encountered in the data set would be a true reflection of agro-dealer s optimization decision. The agro-dealer s market participation and supply of fertilizers can be viewed from two different perspectives. The first incidence is where the agrodealer makes the decision to participate and the quantity to supply simultaneously. In this regard, market participation and supply of fertilizer can be estimated via tobit estimator. Fixed costs associated with market participation do not significantly affect agro-dealers decision to participate in fertilizer market and factors affecting market participation and quantity decisions are one and the same, affecting the dependent variable in the same direction (Ricker- Gilbert and Jayne, 2009). On the other hand, where the agro-dealer makes decision to participate and the quantity to supply sequentially, the double hurdle model to address corner solutions is appropriate. Ricker-Gilbert and Jayne (2009) and Yu (2007) indicate that in the double hurdle model, factors influencing market participation and factors influencing quantity of fertilizer supplied may be different. Thus, the same factors can potentially affect participation and amount supplied differently. Furthermore, Ricker-Gilbert and Jayne (2009) pointed out that fixed costs may affect agro-dealer s decision to participate in the market but they may not affect the quantity supplied. Considering the fact that the quantity of subsidized fertilizer vary over time (Dorward and Chirwa, 2011), it is assumed that agro-dealers make market participation and supply of fertilizer decisions sequentially. To construct a double hurdle model, it is assumed that every agro-dealer is a potential fertilizer trader whose utility function of participating in fertilizer market is assumed as follows: Where.. (1) is a vector of observed socioeconomic and demographic variables which can affect the trader s utility. is a vector of the corresponded coefficients, and is the unobserved heterogeneities for individual and is assumed as. It is assumed that the participation decision is a binary choice for trader. Thus denotes participation in fertilizer market and denotes no participation. Then participation equation becomes: (2) When which implies that trader decide to participate in fertilizer market, a positive supply of fertilizer can be observed. The supply equation is assumed as: Where is a vector of observed supply side factors which can affect trader s quantity of fertilizer supplied,, is a vector of the corresponding coefficients, and and is assumed as. is the unobserved heterogeneities According to Ricker-Gilbert and Jayne (2009), the estimator for the double hurdle model maximizes the following log-likelihood function: Centre for Agric. Res. & Dev, Malawi 54

61 Where represents value of commercial fertilizer supplied, represents the standard normal cumulative density function, represents the standard normal probability density function and represents the standard deviation. Ricker-Gilbert and Jayne (2009) indicate that Cragg s original model assumes that the errors between hurdle 1 and hurdle 2 are independent, normally distributed and the cov. When this assumption is maintained, first the maximum likelihood estimator (MLE) of in the first hurdle measuring participation can be obtained using a probit estimator. Second, the MLE of which represents the parameter for the second hurdle measuring supply, can be estimated from a truncated normal regression model. When equation (4) collapses to the Tobit loglikelihood function. The model specification of the double hurdle estimator can be tested against the Tobit using a likelihood ratio (LR) test. This test can be used to determine whether or not the data supports sequential or simultaneous fertilizer supply decisions. Average partial effects As indicated by Ricker-Gilbert and Jayne (2009), the average partial effects (APE) of the parameter of interest averaged across traders can be estimated after obtaining coefficient estimates. The APE is the partial effect averaged across the population. The first step in obtaining the APE is to derive the partial effect for the explanatory variable of interest (j) for every observation (i) in the dataset. Where the other elements of the independent variables and the unobservable factors are held constant, the APE for double hurdle model is obtained from the following equation: Where are vectors of explanatory variable in hurdle (1) and hurdle (2) respectively? Are the corresponding trader means that are added to the original equations? The parameters to be estimated are. The APE is generally of greater interest than the partial effect at the average of the explanatory variable particularly in nonlinear models and in the case of discrete variables. Dependent variables The dependent variable for the hurdle (1) is a latent (dummy) variable indexing entry in fertilizer market. The dependent variable for the hurdle (2) is value of commercial fertilizer supplied measured in MWK. It is calculated as the quantity of fertilizer a trader supplies (kg) multiplied by the unit market price of fertilizer; hence a good proxy for trader s level of investment in fertilizer market. Explanatory variables The factors that influence traders entry and sale decision are categorized into variables representing asset and liquidity position, risk preferences, relative returns to fertilizer trade, store location, and transaction costs. However, some of the explanatory variables are unobservable (such as ability, liquidity position and risk preference) or difficult to measure empirically (for example, transaction). Hence, proxy variables as observable indicators of unobservable or difficult to measure variables will be used for empirical analysis (Freeman and Kaguongo, 2003). Freeman and Kaguongo (2003) pointed out that for private sector led fertilizer marketing system to be successful; there is need for traders to mobilize the necessary resources for investing in equipment and distribution facilities. The level of traders assets as well as their liquidity position is an indication of ability to participate in fertilizer markets. Store ownership capture the influence of asset on fertilizer decision and total number of full time employees is used as a proxy for firm size (see Table 2 below). It is hypothesized that storeowners are more likely to invest in storage facilities and respond to opportunities for selling fertilizer. Likewise, large firms are expected to have better access to financial, human, and management resources that are necessary for investing in fertilizer trade. These firms are also more likely to benefit from scale economies in retail trade because of their wider distribution and sales network. The ability to fund trading activities is determined by the overall liquidity level position of the trader, which in turn is influenced by whether a trader obtained credit or not. It is hypothesized that lack of access to credit and liquidity constraints are negatively associated with entry and sales decision. Centre for Agric. Res. & Dev, Malawi 55

62 Table 10: Description of Explanatory variables for Double Hurdle Model Variable Type Description Agrochem Binary 1=if trader sale other agrochemicals, 0 otherwise Lkliq Binary 1=if liquidity or access to credit is constraint on fertilizer trade, 0 otherwise Lktrdinf Binary 1=if lack of fertilizer trade information is constraint on fertilizer trade, 0 otherwise. Lksuppl Binary 1=if lack of fertilizer suppliers is a constraint to fertilizer trade, 0 otherwise Subsidy_fertilizer Binary 1=if subsidized fertilizer is a constraint to fertilizer trade, 0 otherwise Storeown Binary Ownership of the store: 1 =trader owns store, 0 otherwise Ownsex Binary Gender of trade: 1=trader is male, 0 otherwise Lkdemand Binary 1=if lack of demand is a constraint on fertilizer trade, 0 otherwise Educlevel Continuous Education of trader: 1=if trader has at least secondary school education, 0 otherwise poplndlg Continuous Population density in store location (persons/km 2 ) totwkrs Continuous Total number of people employed in the store owntrans Continuous Number of vehicle owned for fertilizer trade Advcustm Binary 1=if trader offers technical advice on fertilizer use to customers, 0 otherwise frtptd Binary 1=if trader perceives fertilizer to be more profitable, 0 otherwise prcmgn Continuous Price margin per kg of fertilizer (MWK/kg) Centre for Agric. Res. & Dev, Malawi 56

63 Traders risk preferences are unobserved but their socio-economic characteristics are assumed to closely reflect risk attitude. Therefore, trader s age, gender, level of education attained, and years of experience selling agro-chemicals are included as proxy variables. It is difficult to predict a priori the influence of age on entry decision. Older traders may be more risk averse and therefore, less inclined to invest in fertilizer trade. On the other side, credit market imperfections and reliance on own-capital implies that younger traders are less likely to invest in fertilizer trade because of their smaller capital base. Better educated traders as well as those with experience selling agro-chemicals are hypothesized to have higher management skills and therefore, more likely to accurately assess opportunities for fertilizer trade. These traders are expected to expand trading activities in response to new opportunities. Traders with experience selling agro-chemicals are also hypothesized to have developed contacts with input suppliers that facilitate entry into fertilizer markets. Gender biases in access to resources and opportunities for trade are hypothesized to favor male traders compared to female traders. Hence, the coefficient on the variable for male trader is expected to have a positive influence on entry decision (Freeman and Kaguongo, 2003). It is difficult to predict a priori the influence of FISP on entry and supply decision because of crowding-in or crowding-out effects. Freeman and Kaguongo (2003) indicate that extensive levels of agricultural market segmentation imply that supply conditions and the state of infrastructure in a store s location condition trader s ability to respond to trading opportunities. Poor rural infrastructure raises marketing and distribution cost that is passed on as high farm-gate input cost and lower farm incomes. Collectively, these factors reduce the derived demand for fertilizer and results in smaller quantity of fertilizer sales. Demand conditions are expected to influence the entry and sale decision. Population density in the store location is used to capture potential demand for fertilizer. This data will be obtained from the Malawi 2008 population census and will be measured by the population density in the administrative unit (sub-location) in which a trader s store is located. It is expected that high population density is positively correlated with high level of local demand for fertilizer and favourable trade prospects. Hence, willingness to enter fertilizer market as well as the sales level is expected to increase with rising population density. Some traders are likely to be precluded from entering fertilizer markets because of high transaction costs. Transaction costs are hypothesized to drive large wedges between fertilizer purchase and sales price thereby imposing high entry barriers that make trade unprofitable for several traders. Although these costs are difficult to observe are measured by proxy variables representing whether or not a trader cited lack of access to fertilizer trade information, wholesale suppliers, and limited technical knowledge about fertilizer as important constraints on the entry decision (Freeman and Kaguongo, 2003). Several of these variables are expected to influence both the entry and sales decision while others influence the entry decision but not the sales decision. Additionally explanatory variables capturing investment in transportation facilities, advisory service to farmers, relative returns to fertilizer trading, and price margin will be included in the second hurdle. Investment in transportation provides an important source of competitive advantage and the coefficient is expected to have a positive influence on fertilizer sales. Provision of advisory service to farmers is assumed to be a trade strategy that is expected to expand sales. Relative return in fertilizer trade is an indication of the profitability of fertilizer trading activities. Traders who report higher relative returns in fertilizer trade are expected to have greater incentives to expand trade compared to those who do not. The magnitude of price margins reflects traders ability to reduce unit cost of trading activities. More efficient traders are expected to have lower price margins and the greater incentives to increase sales. The effect of proximity to ADMARC or other government depots on individual input suppliers businesses To estimate the effect of proximity to ADMARC or other government depots distributing subsidized inputs on individual input suppliers businesses, Ordinary Least Squares procedure will be used as shown below. Where Y is the value of annual fertilizer sales MWK, Z is dummy variable measuring private input supplier s proximity to ADMARC or government depots distributing subsidy fertilizers, and S is a vector of supply-side variables (trade sale other agrochemicals, store ownership, population density, number of people employed, number of vehicles, trader offering technical services, and price margin). is the Centre for Agric. Res. & Dev, Malawi 57

64 error term and are parameters. The effect of trader s proximity to ADMARC or other government depots on their businesses is captured by. Data needs The study will use secondary data that will be collected by the Centre for Agricultural Research and Development at the Lilongwe University of Agriculture and Natural Resources under the research project Guiding Investments in Sustainable Agricultural Intensification in Africa (GISAMA). In summary, private input suppliers (both larger dealers like Farmer s World), and smaller scale agro-dealers will be surveyed under the project. Surveying input dealers will reveal the supply side effects. Questions regarding how the private input business has been affected by the input subsidy will be asked. This work will be complimented by surveys of farm households in nearby areas to understand the relationship they have with the input dealers and how they access these markets. GPS coordinates will be collected from households to understand how distance from market and other factors affect input use. Expected output Factors that influence smallholder farmers decision to access the private input market identified. The effect of farmer s proximity to market on their decision to access the private input market determined. The effect of input subsidy program on private input suppliers estimated. The effect of proximity to ADMARC or other government depots distributing subsidized input on individual input suppliers businesses quantified. Thesis Four journal articles References Chirwa E. W. Matita M. and Dorward A. (2010). Targeting Agricultural Input Subsidy Coupons in Malawi. Available from 7 [Date accessed: 15/07/2013]. Chirwa E.W (2005). Adoption of fertiliser and hybrid seeds by smallholder maize farmers in southern Malawi. Available from [Date accessed: 10/07/2013]. Chirwa E.W. and Dorward A. (2012). Private Sector Participation in the Farm Input Subsidy Programme in Malawi, 2006/ /12. Available from [Date accessed: 22/07/2013]. Dorward and Chirwa (2011). The Malawi Agricultural Input Subsidy Programme: to Available from [Date accessed: 12/07/2013]. Fotue L.A.T and Sikod F. (2012): Determinants of the households choice of drinking water source in Cameroon. Available from: Summer2012A/PDF/Determinants%20of%20the %20Households%20Choice.Luc%20Armand%2 0Totouom%20Fotue.pdf. [Date accessed: 09/07/2013]. Freeman H.A and Kaguongo W. (2003). Fertilizer market liberalization and private retail trade in Kenya. Available from 1-s2.0-S main.pdf? [Date accessed 20/08/2013]. Government of Malawi (2012). Malawi Growth and Development Strategy II. Ministry of Economic Planning and Development, Lilongwe. Imperial College London, Wadonda Consult, Michigan State University and Overseas Development Institute (2007). Evaluation of the 2006/07 Agricultural Input Supply Programme, Malawi: Interim Report. Available from ReportFINALXXB.pdf [Date accessed: 13/07/2013]. Kelly V. A. Crawford E. W. and Jayne T.S. (2003). Agricultural input use and market development In Africa: recent perspectives and insights. Available from ( m) [Date accessed: 07/07/2013]. Kelly V. Boughton D. and Lenski N. (2010). Malawi Agricultural Inputs Subsidy: Evaluation of the 2007/08 and 2008/09 Program Input Supply Sector Analysis. Available from ural_inputs_subsidy_evaluation Date accessed: 07/07/2013]. Mabiso A. (2012). Participation of Smallholder Farmers in Biofuel Crop and Land Rentals Markets: Evidence from South Africa. Available from: Centre for Agric. Res. & Dev, Malawi 58

65 /Mabiso_a_2011_paper_1633_abstract_10435_0 %5B1%5D.pdf. [Date accessed: 29/03/2013]. Mason N.M. and Ricker-Gilbert (2012). Disrupting Demand for Commercial Seed: Input Subsidies in Malawi and Zambia. Available from [Date accessed: 11/07/2013]. Morris M. Kelly V.A. Kopicki R.J and Byerlee (2007). Fertilizer Use in African Agriculture: Lessons Learned and Good Practice Guidelines. Available from: [Date accessed: 08/07/2013]. Ricker-Gilbert J. and Jayne T.S. (2009). Do Fertilizer Subsidies Affect the Demand for Commercial Fertilizer? An Example from Malawi. Available from ageconsearch.umn.edu.innopac.up.ac.za/bitstrea m/51606/2/iaae_ref679_jrg_revised.pdf [Date accessed: 11/08/2013]. Seini W. Jones M. Tambi E. and Odularu G. (2011). Input Market Initiatives that Support Innovation Systems in Africa. Available from atives_for_web.pdf. [Date accessed: 09/07/2013]. Simtowe F. and Zeller M. (2006). The Impact of Access to Credit on the Adoption of hybrid maize in Malawi: An Empirical test of an Agricultural Household Model under credit market failure. Available from [Date accessed: 11/07/2013]. World Bank (2005). Jump-Starting Maize Production in Malawi through Universal Starter Packs. Available from e.pdf [Date accessed 09/07/2013]. Xu Z. Burke W.J. Jayne T.S. and Govereh J. (2008). Do input subsidy programs crowd in or crowd out commercial market development? Modeling fertilizer demand in a two-channel marketing system. Available from onlinelibrary.wiley.com.innopac.up.ac.za/doi/ /j x/pdf [Date accessed: 12/08/2013]. Yu X. (2007). Family Structure, Education and Cigarette Smoking of the Adults in China: a Double-Hurdle Model. Available from [Date accessed 09/08/2013]. Takeshima H. Nkonya E. and Deb S. (2012). Impact of fertilizer subsidies on the commercial fertilizer sector in Nigeria: evidence from previous fertilizer subsidy schemes. Available from [Date accessed: 09/07/2013]. Centre for Agric. Res. & Dev, Malawi 59

66 Annex 1.4: Improving maize productivity through the promotion of sustainable and profitable use of fertilizer: the role of complementary, restorative and management practices Joseph S. Kanyamuka, Department of Agricultural and Applied Economics, Lilongwe University of Agriculture and Natural Resources (LUANAR), Bunda College of Agriculture, P. O. Box 219, Lilongwe, Malawi Introduction Background Maize is the main staple crop and remains the dominant crop among smallholder farmers in Malawi. Almost 70 percent of the land among smallholder farmers is devoted to maize cultivation and thus, it is not surprising that maize availability in the country defines the food security situation of the nation. Smallholder agriculture in Malawi has been characterized by low productivity, low technology and labour intensive. Among other factors, the low productivity in smallholder agriculture has been attributed to loss in soil fertility, low application of inorganic fertilizers and traditional low technology rain-fed farming systems (Chirwa et al., 2010, Blackie et al., 1998). Nakhumwa (2004) argued in support that soil erosion and soil nutrient mining through continuous cultivation of crops (mono-cropping), especially staple crops (mainly maize), stands out as the major cause of declining soil fertility and subsequent productivity of soils in Malawi. Further, this is aggravated by increasing population density and subsequent decline in land availability making fallow (or crop rotation), which traditionally used to restore soil fertility and reduce the built up of pests and diseases, almost impossible (Levy, 2005). This has led to degradation of the soil resource base with consequent reduction in yield. However, the food security of resource poor households is critically dependent on the productivity and sustainability of maize based cropping systems (Conroy et al., 2006). Soil fertility depletion in the smallholder farms is the fundamental biophysical root cause for declining per capita food production in sub- Saharan Africa (Kamau et al., 2013, Mutegi et al., 2012). Use of inorganic (also called mineral or commercial) fertilizer and improved crop varieties has received greater attention recently as panacea for smallholder productivity as well as encountering food insecurity. These are being promoted via input subsidies and other campaigns in several African countries (Kamau et al., 2013, Zingore, 2007). The Malawi Agricultural Input Subsidy Programme (MAISP) was first implemented in the 2005/06 agricultural season following a poor harvest season and a high maize import bill to augment domestic supply in 2004/05 agricultural season. The MAISP is largely implemented under government subvention with donor support being in form of overall budgetary support. Achieving sustainable agricultural intensification (and productivity) and food security among most of the poor citizens in many African countries remains the major policy dilemma of many African governments (Sommer et al., 2013). Zerihu et al., (2013) defined sustainable agriculture as the successful management of resources to satisfy changing human needs while conserving natural resources. According to Levy (2005), sustainable agriculture covers technological innovations that promote renewable or environmentally friendly methods to reduce the risk. While the original focus of sustainable agriculture has been on the ecological and environmental aspects, its parameters have grown to include economic, social and political dimensions. Levy (2005) further identified three key challenges for sustainable agriculture in Malawi which are inter-related. These are: 1. Lack of diversity in the farming system which has negative consequences for pest and disease control, family food security and livelihoods and agricultural biodiversity. 2. Intensity of cultivation per unit area, particularly cultivation which has adverse consequences for soil fertility management, food security and livelihoods at household level and, via linkages, within the rural economy as a whole and; 3. Long-term deterioration of soil fertility which has negative consequences for intensity of cultivation and thus for food security and livelihoods. Addressing the food insecurity agenda in Africa hinges on the development of predictable and significant improvement in farming systems using soils which are very low in fertility and subject to the further stress of periodic drought (Conroy et al., 2006). Malawi s smallholder farmers have become increasingly dependent on chemical fertilizers to sustain maize yields. Centre for Agric. Res. & Dev, Malawi 60

67 According to (Sommer et al., 2013), one of the sustainable, intensified nutrient concepts that have proven successful in famer s field is Integrated Soil Fertility Management (ISFM). The other one is Conservation Agriculture (CA). Sustainable agricultural (or ISFM) technologies can go some way towards reducing the need for input use (Kamau et al., 2013). ISFM Africa (undated) defined ISFM as the application of soil fertility management practices, and the knowledge to adapt these to local conditions, which maximize fertilizer and organic resource use efficiency and crop productivity. These practices necessarily include appropriate fertilizer and organic input management in combination with the utilization of improved germplasm. Use of soil amendments including organic materials and mineral fertilizers is highly recommended for the replenishment of soil nutrients, improved soil health and more efficient use of fertilizers in SSA. Addition of organic matter improves nutrient and water retention in soils and increases better synchrony in nutrient supply and crop demand (Kamau et al., 2013). The use of site specific ISFM technologies can also lead to economic benefits if gains in profits due to improved input productivity are able to offset the cost of adoption. This may also result in environment benefits in that efficiency would reduce nitrogen residues in the soil thereby reducing run-off and leaching of nitrates to the environment (Khanna, 1999). Another policy challenge facing African governments is that the adoption of practices needed to restore soil properties and enhance response to inorganic fertilizer remains low. A host of factors are presupposed to suppress economic incentives for adoption of soil fertility management technologies, depending on combination and timing (Kamau et al., 2013). Many sustainable agricultural practices (such as agro-forestry, intercropping and crop rotation) are a package of interrelated components and farmers often adopt pieces of the package rather than the whole and adoption occurs in sequential or stepwise manner. Further, these soil fertility enhancing and management technologies are not sold on shelf as an integrated package and farmers in the components that they assemble. However, missing or underdeveloped markets have also been cited as a reason for limited uptake of recommended practices. Problem Statement and Justification Input intensification was formerly promoted on food crops in the 1970s and 1980s in some parts of Africa through state-led programs featuring subsidized and interlinked credit-input-output marketing arrangements (Govereh et al., 1999). Use of inorganic (also called mineral or commercial) fertilizer and improved crop seed varieties has received greater attention recently as panacea for smallholder productivity as well as encountering food insecurity. These are being promoted via input subsidies and other campaigns in several African countries (Kamau et al., 2013, Zingore, 2007). Just like the universal starter pack programme, the Malawi Agricultural Input Subsidy Programme (MAISP) is being promoted blindly with an immediate goal of increasing crop production and enhancing food security. The designers of both the two policy interventions paid little, if no attention to the objective of sustainable agriculture. However, one of the major ironies emanating from applied research is that the greater use of modern inputs such as fertilizer appears to be marginally profitable or even unprofitable to use on the main staple food crops in the region. There is an increasing realization that input and technology intensification alone to raise farm productivity growth is likely to be impeded without paying greater attention to complementary management practices (suitable to local conditions) that enhance soil fertility and promotes the efficiency use of mineral fertilizers and associated improved seeds. Nevertheless, the efforts in the promotion of these soil fertility enhancement technologies have ignored soil and climatic (agro-ecological) variations found in smallholder farming areas hence the limited resultant adoption. Blackie et al. (2005) asserted that such technologies are either incompatible with the local conditions and farmer resources (which are severely limited), or inefficient thereby affecting the profitability of fertilizer. Fertilizer response varies as does the marginal productivity of nitrogen according to agro-ecological and soil conditions, both among and within farms. Further, Zingore (2007) noted by the impact of the efforts will be very limited, unless the fundamental issues of providing the crops with balanced and adequate nutrients are addressed, and fertiliser recommendations are fine-tuned to account for highly variable soil fertility conditions. The researcher agrees with (Sommer et al., 2013) who suggested that the optimal technical approach of SSA countries to increase the use of mineral fertilizer is likely to be a function of a number of location-specific, agro-climatic, demography and economic variables. Therefore, this research will be of paramount importance in filling the knowledge vacuum by firstly, identifying the best bet complementary, Centre for Agric. Res. & Dev, Malawi 61

68 restorative and soil fertility management (ISFM) technologies that are suitable to each of the four agro-ecological zones in Malawi and to farmer situations where they will make a positive impact. Secondly, the study will go a long way in providing answers as to why some technologies exhibit dismal performance in farmers fields after demonstrating excellent performance under research field trials within the homogenous environmental conditions. Thirdly, the study will help in identifying factors that go into the choice process of the sustainable agricultural technologies. Understanding how factors leading to adoption vary by location and type of technology may help technology promoters know who to target in diverse settings (Rubas, 2004). Thus overall, the study will provide fundamental guidance to the design of an appropriate and relevant policy package of technologies for improving maize productivity and subsequent achievement of sustainable food security in the country. Objectives Overall Objective The overall objective of this research is to identify the strategies to promote the sustainable and profitable use of fertilizer and improved seed use among smallholder maize farmers. Specific Objectives Specifically, the study seeks to achieve the following objectives: 1. To identify complementary and restorative management practices to the expansion of fertilizer and improved seed use that can raise maize productivity among smallholder farmers in the four agro-ecological zones in Malawi. 2. To determine key factors that influence farmers decisions on incidence and extent of adoption of ISFM (complementary and restorative management) technologies among smallholder farmers in maize-based farming systems in Malawi. 3. To evaluate maize productivity and fertilizer use efficiency among smallholder farmers in Malawi 4. To provide appropriate and relevant policy guidance to the promotion of fertilizer input intensification strategies that can raise maize productivity and subsequent achievement in food security. Research Questions The project will address the following main research questions: 1. Do the complementary and restorative management practices to fertilizer expansion and improved seed use have any bearing of improving maize productivity among smallholder farmers? 2. Which fertiliser application rates are optimum in maize in the four agroecological zones in Malawi given the best bet sustainable agricultural technologies? 3. Is maize land productivity higher for adopters of ISFM technologies after controlling for differences in fertilizer use intensity? If yes, how much? METHODOLOGY Study Area The study will be conducted in the four agroecological zone districts in Malawi which are major producers of maize and with high agricultural potential. Malawi has four major agro-ecological zones categorised based on soil types, vegetation types and climatic conditions, namely: High altitude plateau, medium altitude plateau, Lake shore plain and Shire Valley (Saka et al., 2006). Data Collection and Sampling Techniques The research will adopt an experimental approach to potentially control for confounding unobservable factors that may influence adoption of input intensification strategies. A two-stage stratified sampling method will be employed. Within the agro-ecological zones (smallholder maize production systems), districts will be selected based on high inclusivity of maize crop. An extension planning area (APA) will purposively be selected according to the same criterion of high inclusivity of the maize crop. Data will be collected both at household and plot level. Individual farmers will be randomly selected from the EPA that will be followed through the 5 months period of maize cultivation. Field activities will be monitored and recorded on a weekly or monthly basis. This will involve collecting data on management activities such time of weeding, fertilizer application and implementation as well as performance of ISFM technologies, among others. A semi-structured questionnaire will also be administered to farmers through a survey. Centre for Agric. Res. & Dev, Malawi 62

69 Sample Size The sample size will be determined using the formula below (Edriss, 2003): factors affecting maize production. According to Salvatore (2003), the Cobb-Douglas production function is the simplest and most widely used production function in empirical work today. Further, the Cobb-Douglas production function has been chosen because it is easy to calculate factor elasticities. The formula for the Cobb- Douglas production function is given by: Where: Sample size Value yielding the desired degree of confidence p Therefore, An estimate of the population (0.5) Absolute size of the error in estimating Therefore, the sample size will be 384. Accounting for non-response error (10%), the sample size will be adjusted upwards to 422. This sample will be divided among the agroecological zone districts to be studied based on proportional to size sampling (PPS) Data Analysis The collected data will be subjected to data cleaning and other management issues prior to analysis. The data will be analysed using Microsoft Excel, Statistical Package for Social Scientists (SPSS) and Stata. Descriptive statistics such as means, percentages and frequencies will computed to explain some socio-economic characteristics of the sample. These will be subjected to statistical tests to see if there will be any significant differences in the socio-economic characteristics between adopters and nonmembers of ISFM technologies. Analytical Techniques The Cobb-Douglas Production Function The Cobb Douglas Production Function will be used to relate maize output (kg/ha) to the set observed inputs such as different application levels of fertilizer so as to identify the main Where Q= output in physical units, L = quantity of labour, K = quantity of capital, and, α and β are positive parameters estimated in each case from the data. The parameter α refers to the percentage increase in Q for a 1 percent increase in L, while holding K constant. Thus α is the output elasticity of labour. Similarly the parameter β refers to the percentage increase in Q for a 1 percent increase in K, while holding L constant. Thus β is the output elasticity of capital. The sum of α and β indicates the nature of returns to scale. When, indicates the existence of constant returns to scale, that is, tripling the inputs will triple the output; indicates decreasing returns to scale, doubling the inputs will less than double the output; and indicates increasing returns to scale, that is, doubling the inputs, will more than double the output. In its stochastic form, the Cobb-Douglas production function is represented as: Where: Y is maize output (yield/ha) E is the base of the natural logarithm Is the stochastic disturbance term Is the vector of variable resources? However, to allow for the use of the Ordinary Least Squares (OLS) estimation method in running a production function regression, the Cobb-Douglas function needs to be log transformed (Edriss et al., 2005, Tung et.al., 2005 and Gujarati, 1998). The logarithmic transformation of the model is: The model is linear in parameters and E and thus will be estimated using ordinary least squares method. The variables to be run in the model will included both quantitative and qualitative. These are land size (ha), labour (man days), seed rate (kg/ha), fertilizer level (kg), ISFM technology use, manure, access to agricultural extension, age Centre for Agric. Res. & Dev, Malawi 63

70 of the household head and gender of the household head. The Tobit model Chiputwa et al., (2011) defines adoption as the decision of a farmer to use a particular sustainable agricultural technology component. With regard to sustainable farming practices, statistical challenges of modelling adoption decisions involving packages or bundles of inputs have been addressed in a number of ways over the past few decades. Several recent studies about adoption of soil fertility management practices in Eastern and Southern Africa have used a series of single probit or logit equations to model the range of practices independently (Kamau et al., 2013). However, most adoption studies are based on censored data and one of the mostly applicable models is the tobit model. Proposed by James Tobin (1958), a tobit model is a statistical model to describe the relationship between a nonnegative dependent variable and an independent variable (or vector). Adoptions of technological innovations in general is not a time decision but rather a step-wise decision made after weighing carefully opportunity costs at each point. This study will therefore attempt to separate, firstly, factors influencing the decision on whether or not to adopt the sustainable agricultural technology. Secondly, the study will identify factors determining the intensity (extent) of use of the technology. A Tobit model is thus, preferred to logit and probit models in that tobit models measure not only the probability of adoption but also the intensity of use of the technology once adopted (Waithaka et al., 2007). However, estimating the impact of adoption for each of the subgroups of adopters and nonadopters of one or both component technologies separately could lead to a self-selection bias. This bias arises because farmer s endogenously self -select themselves into these sub groups and observed and unobserved characteristics of farmers may influence both their self-selection into adopters and non-adopters and the productivity enhancing outcome of that decision (Khanna, 1999). Heckman s two-stage selection model will therefore be used to minimise sample selection bias as the adoption decision of fertilizer and complementary management practices by the farmers may not be random. The Tobit specification with left censoring at zero: m Y * i = ß 0 + i 1 ßiX i +U i, i = 1, 2... m; Where Y= Y*, if Y* > 0, Y = 0 if Y* < 0 and Y= max (Y*,0) Where Y * i = the proportion of the crop area a farmer commits to a particular sustainable agricultural technology (dependent variable) ß 0 = an intercept ß i = coefficients of i th independent variable X i = independent variable, and 'i' is 1, 2, 3 m U i = unobserved disturbance term According to Zingore (2007), three categories of soil fertility management practices are: (i) inorganic fertilizers, (ii) other soil amendments and; (iii) erosion control. Thus the sustainable agricultural techchnologies which will be considered in this research include: organic manure (farmyard, green manure and compost), maize-legume intercropping (involving soybeans, cowpeas, pigeon peas, groundnuts, etc), agro-forestry, crop rotation and soil erosion control measures such as contour ridging, contour bunds and planting of vertiver/elephant grass which is known to reduce run-off. A number of socio-economic factors assumed to influence both the incidence and intensity of adoption of the ISFM technologies will be considered as explanatory variables. These include farm size, labour availability, asset status (plus livestock ownership), education level of the household head, contact with extension agent, maize yield level, age of the household head, gender of the household head, land tenure system, experience of the farmer and others. The Tobit model parameters are estimated by maximizing the Tobit likelihood function of the f form: Y β X - i i i i L (2) y* 0 1 f δ y* 0 Where f and F are respectively, the density function and cumulative distribution function of Y i* yi*>0 means the proportional area over those i for which y i*>0, and y i* 0 means the proportional area over those i for which y i* 0. The marginal effect of an explanatory variable on the expected value of the dependent variable is: i X E Y i F z i F (3) Centre for Agric. Res. & Dev, Malawi 64

71 Where, Maddala, (1997) i X i is denoted by z, following The change in the probability of technology adoption as independent variable X i changes: F X z i f z i (4) The change in intensity of technology use with respect to a change in an explanatory variable among farmers: E( Y / Y 0) i i X * i i = f ( z) f ( z) 1 Z F( z) F( z) 2 (5) Where, F (z) is the Cumulative Normal Distribution of z, f (z) is the value of the derivative of the normal curve at a given point (i.e., unit normal density), z is the Z score for the area under normal curve, i is a vector of Tobit Maximum Likelihood estimates and σ is the standard error. Normality or homoskedasticity fail to hold, the Tobit model may be meaningless. In OLS, estimates are consistent but not efficient when the disturbances are heteroscedastic. In the case of the limited dependent variable models also, if we ignore hetroscedasticity, the result estimates are not even consistent i.e. is the regression coefficient is upward biased (Maddala, 1997). Estimation of the whole function would give more efficient estimates, but excluding inconsistencies or biases. In this context, the dependent variable of the technology function is censored by unobservable latent variable influencing the decision of whether or not to adopt the technology. The assumption underlying a Tobit estimation is that farmers are unconstrained which is untenable in light of the fact that proportional area under technology is below the saturation point. Hence, it is necessary to use the Heckman selection model to account for sample selection bias (Greene, 2000). The Heckman s sample selection model where a Probit model for the participation or selection equation is estimated and a regression model, which is corrected for selectivity bias, is specified to account for the proportional of the area committed to a particular technology. To employ Heckman s sample selection, first, the probability of participation will be modelled by Maximum Likelihood Probit, from which inverse Mill s ratio will be estimated. In the second-stage, the estimated Inverse Mill s Ratio (IMR) will be included as right-hand variable in the corresponding technology intensity use function. The Probit model is specified as: i = 1... n (6) Where: Y i is a dummy variable indicating the technology adoption that is related to it as Y i = 1 if Y i > 0, otherwise Y i = 0 i are the variables determining participation in the Probit model, Xi is unknown parameter to be estimated in the probit regression model, i is random error term Then the parameters can consistently be estimated by OLS over n observations reporting values for Y i by including an estimate of the inverse Mill s Ratio, denoting i, as an additional regressor in (2). More precisely selection model is specified as: Y i = x i I + I + I (7) Where: Y i is the proportional of the area under the ISFM technology in question in the secondstep, i is unknown parameter to be estimated, Xi is the explanatory variables determining proportional of the area under the ISFM technology i is a parameter that shows the impact of adoption of the technology, is the error term The probity model will therefore precede the two bit model. The two bit model will compute variables that influence the farmer s decision on intensity of use of the technology. TATA econometric Software will be employed to run the models (Tobit and Heckman two-stage selection). Determination of Fertilizer Use Efficiency Centre for Agric. Res. & Dev, Malawi 65

72 The level of fertiliser use efficiency (profitability) will be estimated by comparing marginal value product (MVP) with Marginal Factor Costs (MFC) fertilizer. The ratio, R = MVP/MFC will be calculated and used to estimate the level of economic efficiency in use of different fertilizer levels on maize as applied by Abed et.al., (2011). The general equation to find the optimum fertilizer usage that was to maximize profit is given below: (1) And optimal levels of the inputs are given below by simultaneous equations of the partial derivatives: (2) Where: T = calculated t-value X 1 and X 0 are means for the two categories; adopters and non-adopters, in this case. SE( X X ) 1 0 [ n X n X ][ n (1 X ) n (1 X )] n n ( n n ) n 1 and n 0 are the sample sizes for the two categories. The following null (H ow) and alternate (H a) hypotheses will be tested: H o : x x ; H x x 1 0 a : Where: x 1= mean for ISFM technology adopter (3) x 0 = mean for ISFM technology nonadopter Where: π = net profit (MK/ ha) (4) REFERENCES Abide, M., Afar, M., Qudus, A., Tahir2, A., and Fatima, N. (2011). Resource Use Efficiency Analysis of Bt Cotton Farmers in Punjab, Pakistan. Pak. J. Agri. Sci, 48 (1), y = quantity of maize X1, x 2... xn = levels of inputs X1, = Labour in man days; x 2=Land in hectares, X 3=Seed in kg per hectare, X 4= Fertilizer used in production of maize measured in kilograms. p = value of maize (MK/kg) w 1, w2... wn = wages of man hours worked (MK/day) and costs of inputs in production of maize. Evaluating Maize Productivity between ISFM Technology Adopters and Non Adopters To achieve this objective, a t-statistic will be used to test for the significant difference in maize productivity across a range of ISFM technologies between adopters and non - adopters. In general, t-test for mean difference is used to assess if means from two sub-groups are significantly different or not. It is given by the following formula: Blackie, M., and Mann, C. (2005). The Origin and Concept of the Starter Pack. In: Calibre Consultants (2005). Starter Packs: A strategy to Fight Hunger in Developing Countries? Lessons from the Malawi Experience Ed (Levy, S.). CABI publishing Blackie, M., Benson, T., Conroy, A., Gilbert, R., Kanyama-Phiri, G., Kumwenda, j., Mann, C., Mughogho, S., Phiri, A. and Waddington, S. (1988). Malawi: Soil Fertility Issues and Options-a discussion paper. MPTF/Ministry of Agriculture and irrigation, Malawi Calibre Consultants (2005). Starter Packs: A strategy to Fight Hunger in Developing Countries: Lessons from the Malawi Experience Ed (Levy, S.). CABI publishing Chiputwa, B., Langyintuo, A.S., and Wall, P. (2011). Adoption of Conservation Agriculture Technologies by Smallholder Farmers in the Shamva District of Zimbabwe: A tobit Application Centre for Agric. Res. & Dev, Malawi 66

73 Chirwa, E.W., Mhoni, V., Kachule, R., Chinsinga, B., Musopole, E., Makwenda, B., Masankhidwe, C., Kalumula, W., and Kankangadza, C. (2010). The Malawi Agricultural Input Subsidy Programme: Lessons from Research Findings, Future Agricultures Consortium (FAC) Conroy, A., and Malcom, B. (2006). The Collapse of Agriculture. In: Conroy, C. A., Malcolm, J. B., Whiteside, A., Malewezi, C.J., and Sachs, D.J. (2006) Poverty, AIDS and Hunger: Breaking the Poverty Gap in Malawi. New York: Palgrave Macmillan Salvatore, D., & Reagan, D. (2011). Shaum's Outline of Statistics and Econometrics. 2 nd edition. McGraw-Hill. Edriss, A. K. (2003). A Passport to Research Methods: Research Skills Building Approach. Las Vegas: International Publishers and Press. Govereh, J., T.S. Jayne, and Nyoro, J. (1999). Smallholder Commercialization, Interlinked Markets and Food Crop Productivity: Cross-Country Evidence in Eastern and Southern Africa Kamau, M., Smale, M., and Mutua, M. (2013). Farmer Demand for Soil Fertility Management Practices in Kenya s Grain Basket. Selected Paper Prepared for Presentation at the Agricultural and Applied Economics Association (AAEA) and CAES joint meeting Washington, D.C, USA Khanna, M. (1999). Sequential Adoption of Site Specific Technologies and its Implications: Double selectivity Model. Department of Agricultural and Consumer Economics, University of Illinois, USA Mangison, J.H. (1999). Land Degradation, Profitability and Diffusion of Erosion Control Technologies in Malawi. PhD thesis Nakhumwa, T.O. (2004). Dynamic Costs of Soil Degradation and Determinants of Adoption of Soil Conservation Technologies by Smallholder Farmers in Malawi. PhD thesis. University of Pretoria. South Africa. Sommer, R., Bossio, D., Dester, L., Dimes, J., Kihara, J., Koala, S., Mango, N., Rodriguez, D., Thierfelder, C., and Winowiecki, L. (2013). Profitable and Sustainable Nutrient Management Systems for East and Southern African Smallholder farming Systems-Challenges and Opportunities: A synthesis of the Southern and eastern African Situation in Terms of the Past Experiences, Present and Future Opportunities in Promoting Nutrient Use in Africa Waithaka, M., Thornton, P., Shepherd, K., and Ndiwa, N. (2007). Factors affecting the use of Fertilizers and Manure by smallholders: the Case of Vihiga, Western Kenya. Springer Science plus Business Media. Zingore, S. (2007). Maize Productivity and Response to Fertilizer Use as Affected by Soil Fertility Variability, Manure Application, and Cropping System. Springer Saka, A.R., Mtukuso, A.P., Daudi, A.T., Banda M.H.P., Mwenda, A.R.E., and Phiri, I.M.G, (2006). A Description of Agricultural Technologies used by Farmers in Malawi. Departmnt of Agricultural Research Services in Malawi, Lilongwe. Centre for Agric. Res. & Dev, Malawi 67

74 Annex 1.5: Profitability of Fertilizer Use: Evidence from Malawi Francis Darko, Department of Agricultural Economics, Purdue University, 403 W. State Street, West Lafayette, USA, IN Introduction Improvement in agricultural productivity is widely regarded as a major channel through which the widespread food insecurity and poverty in Africa can be curtailed and even eradicated (Future Agricultures, 2010). This notion is based on the well-known fact that majority of the poor and food insecure in Africa derive their livelihood from agriculture. Unfortunately, agricultural productivity has been very low since the 1960s - average annual growth in agricultural productivity is less than 1% for the continent as a whole, and negative for some subregions (FAO statistic). Malawi typifies the agricultural productivity situation in Africa. For the past two decades, the productivity of most agricultural crops in the country has increased only modestly. The already modest increase in productivity is further undermined by population growth (MoAFS, 2011). The Ministry of Agriculture and Food Security (2011) estimates that the country s yield gap 17 ranges from 38% to 53% for cereals, and 40% to 75% for legumes, implying that there is enough room for improvement in productivity. Such an improvement will have important implications for food security and poverty alleviation because majority of the poor and food insecure households engage predominantly in agriculture. A major factor accounting for the low agricultural productivity among Malawian smallholders, who produce about 80% of the country s agricultural output, is limited utilization of modern inputs, particularly inorganic fertilizer 18. This is so because, rampant depletion of nutrients that results from soil degradation and soil mining on the part of farmers has rendered majority of the nation s arable land poor in soil nutrients. As in many other African countries, the low level of fertilizer use emanates from the fact that the adoption fertilizer appears to be unprofitable or only marginally profitable. Hence in order to promote the use of inorganic fertilizer in Malawi, it is imperative that the profitability of fertilizer is enhanced. This will not only encourage the use of fertilizer but will also maximize the benefits derived from the large-scale farm input subsidy program that the government is currently implementing. The goal of this study is therefore to identify ways through which the profitability of fertilizer use can be improved in order to promote the use of fertilizer, and increase the benefits of the input subsidy program. Specifically the study seeks to identify ways through which the response rate of fertilizer can be improved, and means through which the cost and risk in input and output markets systems reduced. The study focuses on maize because apart from being the main staple crop of the country, it is grown by about 90% of farmers and accounts for over 70% of all the arable land under production (Sauer, 2009). Moreover, maize is one of the primary focuses of the input subsidy program. The following questions will be addressed: 1. What is the yield response rate of fertilizer in the production of maize, and how does it vary over space? 2. What are the factors that account for spatial variation in yield response rate? 3. Where is fertilizer profitable and where is it not? 4. How can the cost and risk involved in the fertilizer and maize markets be reduced in order to improve upon the profitability of fertilizer? The following hypothesis will also be tested 1. The input subsidy program has a significant effect on fertilizer profitability Conceptual Framework The conceptual framework of this study is based on a yield response model expressed as: 17 Yield gap is the difference between potential yield and the yield of the average farmer (Lobell et al., 2009). 18 Other factors affecting agricultural productivity include unfavourable weather conditions, declining soil fertility, limited adoption of sustainable land management practices, limited agricultural extension services, market failures, limited access to agricultural credit and (input and output markets), and infrastructural related problems (World Bank, 2007; Kilic et al., 2013; MoAFS, 2011). Where (1) a total kilogram of maize per acre of land is that household j obtained from plot i. Is a vector of growth inputs applied to plot i by household j. Growth inputs consist of such inputs as seed, nutrients, and water that are directly involved in the biological process of plant growth and development (Guan et al. 2006; Xu et al., 2009). Is a vector of Centre for Agric. Res. & Dev, Malawi 68

75 factor sand practices that are not directly involved in the growth and development process of plants but influence the response rate of plants to the growth inputs. Factors such as soil conditions and pesticides application; and practices such as weeding, intercropping and tillage are considered in this category. H consists of other inputs such as labor and capital that are important for crop production. Fertilizer application in maize production is typically done twice per planting season in Malawi. Basal fertilizer (NPK 23:21:0 + 4S) is applied within a week after planting; and is followed by top dressing, mainly Urea (NPK: 46:0:0) or Calcium Ammonium Nitrate (CAN), 21 days later. Because basal fertilizer is entirely nitrogen and phosphorus, Urea is entirely nitrogen, and CAN is predominantly nitrogen, the nutrients to be considered in the study will include only nitrogen and phosphorus. Farming, like other human activities, occur within a social system that is characterized by a network of interdependencies among actors. Farmers, in developing countries especially, adapt their behaviors to the behavior of other farmers in their communities through communication and/or comparison. This is particularly true in the area of adoption and utilization of improved farm inputs. This study therefore believes that, a farmer s decision regarding fertilizer application - whether or not to apply inorganic fertilizer, the kind and amount of fertilizer to apply, and whether or not to apply basal and/or top dressing fertilizers- is partly influenced by the opinions and behaviors of other farmers. Accordingly, the profitability of fertilizer application of a particular farmer will depend on the profitability of other farmers. That notwithstanding, previous studies on fertilizer profitability did not account for this interdependencies. Considering the interdependencies between farmers, the profitability of fertilizer application is expressed as: The consideration of soil conditions in this study is informed by the extant agronomic literature which suggests that soil conditions affect crop yield in a couple of ways. Firstly, the availability of soil nutrients, particularly phosphorus, to plants is known to be hampered by soil acidity (Burke et al., forthcoming). Secondly, different soil types have different nutrient composition and water holding capacities. Loamy soils for instance are known to contain more organic matter (hence more nitrogen and soil microbes) than sandy and clayey soils; while clayey soils are also known to have relatively higher water holding capacity than sandy and loamy soils. Empirical Model Equation (1) is represented with a conditional yield response model, which is given by: Where: (2) = is total kilograms of maize per acre of land that household j obtained from plot i = is the quantity of nutrient n (nitrogen or phosphorus) in total inorganic fertilizer applied by household i on plot j. = is a vector of factors such as soil type, soil acidity, whether or not plot was tilled, whether or not maize was intercropped with leguminous crop, and number of times weeds were controlled by household i on plot j. = hours of labor (hired and family labor) used by Where is the profitability of nutrient n applied by household i; is the profitability of nutrient n applied by the neighbors of household i; is a vector of soil conditions such as soil type and soil acidity of the maize plots of household i; is a vector of household characteristics; and is a vector of supply side factors( e.g. road accessibility, nearness to fertilizer dealers and the time of application)that affect the profitability of fertilizer application. These supply side factors have not been controlled for in the existing fertilizer profitability literature. household i on plot j. = error term. Conditional yield response model is specified because, compared to the von Liebig s function, it provides a more flexible relationship between yield and inputs (Guan et al., 2006; Xu et al., 2009 and Burke et al., forthcoming). From equation (2), the average and marginal products of nitrogen and phosphorus ( respectively) will be derived; and following Xu et al. (2009), the marginal and average value cost ratios computed as follows: Centre for Agric. Res. & Dev, Malawi 69

76 (3) (4) Where and are, respectively, the average and marginal cost value cost ratios of nutrient n(measures of fertilizer profitability); is the expected per kg price of maize; expected per kg price of nutrient n, and is the and are average and marginal products of nutrient n respectively. Equations (3) and (4) will be estimated for, and compared across, the various regions and districts of Malawi. The determinants of and will then be identified with a spatial econometric model. Spatial econometrics is used in order to account for the interdependences that exist between farmers. The model is specified as follows: Centre for Agric. Res. & Dev, Malawi 70

77 = the number of farming seasons farmer applied fertilizer = dummy variable for top dressing application, = 1 if only top dressing was applied (5) And (6) Where: = weight matrix measuring the interdependences between farmers = The type of soil on maize plot = The level of acidity in the soil = whether or not soil is tilled during land preparation = whether or not the road to the fertilizer dealer is paved = Distance to the closest fertilizer dealer = dummy variable for basal fertilizer application, = 1 if only basal fertilizer was applied = dummy variable for choice of fertilizer for top dressing, = 1 for CAN, = 0 for Urea References = error term Lobell, D. B., Cassman, K.G. and Field, C.B Crop Yield Gaps: The Importance, Magnitudes, and Causes. The Annual Reviews of Environmental and Resources. 34: Hardy, T Malawi: Soil fertility Issues and Options. A discussion paper. Sauer, J The Economics of soil fertility Management in Malawi. Applied Economics Perspective Policy 31 (3): applied on time = whether or not fertilizer was Centre for Agric. Res. & Dev, Malawi 71